Reinventing Identities: The Gendered Self in Discourse (Language and Gender Series)

Citation preview

Reinventing Identities: The Gendered Self in Discourse

Mary Bucholtz A.C. Liang Laurel A. Sutton, Editors

OXFORD UNIVERSITY PRESS

REINVENTING IDENTITIES

STUDIES IN LANGUAGE AND GENDER Mary Bucholt/,, General Editor Advisory Board Penelope Eckcrt, Stanford University Kira Hall, Yale University Janet Holmes, Victoria University of Wellington, New Zealand Miyako Inoue, Stanford University Don Kulick, University of Stockholm Sally McConnell-Ginet, Cornell University Marcyliena Morgan, University of California, Los Angeles/Harvard University Deborah Tannen, Georgetown University Ana Celia Zentella, Hunter College, City University of New York

Reinventing Identities: The Gendered Self in Discourse Edited by Mary Bucholt/,, A. C. Liang, and Laurel A. Sutton

REINVENTING

IDENTITIES The Gendered Self in Discourse

Edited by Mary Bucholtz A. C. Liang Laurel A. Sutton

New York

Oxford

OXFORD U N I V E R S I T Y PRESS

1999

Oxford University Press Oxford New York Athens Auckland Bangkok Bogota Buenos Aires Calcutta Cape Town Chennai Dar es Salaam Delhi Florence Hong Kong Istanbul Karachi Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi Paris Sao Paulo Singapore Taipei Tokyo Toronto Warsaw and associated companies in Berlin Ibadan

Copyright © 1999 by Oxford University Press Published by Oxford University Press, Inc. 198 Madison Avenue, New York, New York 10016 Oxford is a registered trademark of Oxford University Press, Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Library of Congress Calaloging-in-Publication Data Reinventing identities : the gendered self in discourse / edited by Mary Bucholtz, A. C. Liang, and Laurel A. Sutton. p. cm. — (Studies in Language and Gender : 1) Includes bibliographical references and index. ISBN 0-19-512629-7; ISBN 0-19-512630-0 (pbk.) 1. Language and languages—Sex differences. 2. Gender identity. T. Ruchnltz, Mary, 1966- . II. Liang, A. C. III. Sutton, Laurel A. IV. Series. P120.S48R47 1999 306.44—dc21 98-50041

9 8 7 6 5 4 3 2 1

Printed in the United States of America on acid-free paper

To the memory of D. Letticia Galindo (1952-1998), whose pioneering work has helped to reinvent language and gender studies

This page intentionally left blank

SERIES FOREWORD

In the past decade, the subfield of linguistics known as language and gender studies has undergone an intellectual renaissance. From its original concern with sexist language in the 1970s to the 1980s' debate over "difference" and "dominance" models, language and gender research in the 1990s has developed its links to feminist and social theory and expanded its scope to include the interaction of gender with race, ethnicity, sexuality, social class, nationality, and other dimensions of social identity. This new body of work builds on the foundations of earlier research while integrating key insights of recent theory. Such research has interdisciplinary relevance: current scholarship addresses a wide audience, offering insights not only to linguists but also to researchers in anthropology, sociology, ethnic studies, gender studies, and related disciplines. Yet, despite the topic's enduring interdisciplinary interest and the outpouring of important new scholarship, there has never been a series devoted to language and gender research until now. Oxford's series Studies in Language and Gender fills this gap by offering a broadbased interdisciplinary forum for the best new scholarship in the field. The mandate of the series is to encourage innovative work on language and gender, a goal that may be achieved through the revisitation of familiar topics from fresh vantage points, through the introduction of new avenues of research, or through new theoretical or methodological frameworks. This emphasis ensures that the rejuvenated field of language and gender will continue to be replenished with original research. The series is also interdisciplinary in its scope: volumes may be authored by scholars in such disciplines as anthropology, sociology, literary studies, education, psychology, ethnic studies, and women's studies, as well as linguistics. As the inaugural volume of Studies in Language and Gender, Reinventing Identities offers a broad vision of what the field of language and gender studies will look

VIII

SERIES FOREWORD

like in the new millennium. Reinventing Identities is as wide-ranging as gender itself, which takes on new and surprising forms in new contexts. The volume emerges from what might be called the "third wave" of language and gender scholarship. The goal of this new approach is to understand the diversity of gendered experiences as they play out in a variety of situations. Third-wave language and gender research makes explicit its connections to feminist theory; of particular significance are those constructionist perpectives that emphasize how gender identities and ideologies are achieved in discourse. But this approach does not examine language to the exclusion of other social practices, such as physical self-presentation, gesture and movement, and activities. Such details are crucial for arriving at specific, local forms of gender, in contrast to approaches that aim for a general description of "women's use of language." Reinventing Identities counters this well-intentioned but reductive strategy with a series of studies of gender on the ground, formed under conditions of community and contact, shaped moment by moment through the details of discourse. The fluidity of gender illustrated by the chapters of this volume suggests an alternative to more totalizing frameworks, an alternative that respects the variety of gendered selves that discourse makes possible. Reinventing Identities attends to myriad cultural forms of gender: within the U.S. context, chapters focus variously on African Americans, Latinos, Native Americans, and European Americans; on a more global scale, contributors examine discursive gender relations in local contexts in Europe and Africa, as well as in North America. And via the influence of the emergent field of queer linguistics, Reinventing Identities includes a sizable number of studies of sexuality as well as gender, the first volume of its kind to have substantial representation of both fields of inquiry. A contextually and theoretically rich collection of studies of the gendering, ungendering, and regendering of language, Reinventing Identities is an important contribution to the field's current reinvention of itself. The volume invites scholars and students alike to rethink what it means to study the intersection of language and gender and where that intersection is located. The answers offered in its many chapters are as diverse, diffuse, and dispersed as the gendered selves who populate these pages. This undoing of a single unified tale of language and gender is the first step to envisioning new forms of feminist scholarship within linguistics. —Mary Bucholtz, Series Editor

The Reinventing Identities Web site, featuring additional data, graphics, and audio and video clips from the studies in this book, can be found at http: //www-english.tamu.edu/ pers/fac/bucholt/7oslg/re-id

CONTENTS

Contributors, xiii

Bad Examples: Transgression and Progress in Language and Gender Studies, 3 Mary Bucholtz

Part I 1

Identity as Invention

No Woman, No Cry: Claiming African American Women's Place, 27 Marcyliena Morgan

2 Coherent Identities amid Heterosexist Ideologies: Deaf and Hearing Lesbian Coming-Out Stories, 46 Kathleen M. Wood 3 Good Guys and "Bad" Girls: Identity Construction by Latina and Latino Student Writers, 64 Marjorie Faulstich Orellana 4 Constructing the Irrational Woman: Narrative Interaction and Agoraphobic Identity, 83 Lisa Capps 5 Contextualizing the Exotic Few: Gender Dichotomies in Lakhota, 101 Sara Trechter

x

CONTENTS

Part 11 Identity as Ideology 6

Changing Femininities: The Talk of Teenage Girls, 123 Jennifer Coates

1

Rebaking the Pie: The WOMAN AS DESSERT Metaphor, 145 Caitlin Mines

8

All Media Are Created Equal: Do-U-Yourself Identity in Alternative Publishing, 163 Laurel A. Sutton

9

Strong Language, Strong Actions: Native American Women Writing against Federal Authority, 181 Rebecca J. Dobkins

10 "Opening the Door of Paradise a Cubit": Educated Tunisian Women, Embodied Linguistic Practice, and Theories of Language and Gender, 200 Keith Walters

Part III 11

Identity as Ingenuity

The Display of (Gendered) Identities in Talk at Work, 221 Deborah Tannen

12 Gender, Context, and the Narrative Construction of Identity: Rethinking Models of "Women's Narrative," 241 Patricia E. Sawin 13 Language, Socialization, and Silence in Gay Adolescence, 259 William Leap 14 Turn-Initial No: Collaborative Opposition among Latina Adolescents, 273 Norma Mendoz.a-De.nton 15 Conversationally Implicating Lesbian and Gay Identity, 293 A. C. Liana

Part IV

Identity as Improvisation

16 Indexing Polyphonous Identity in the Speech of African American Drag Queens, 313 Rusty Barrett 17 "She Sired Six Children": Feminist Experiments with Linguistic Gender, 332 Anna Livia

CONTENTS

18 Purchasing Power: The Gender and Class Imaginary on the Shopping Channel, 348 Mary Bucholtz 19 From Folklore to "News at 6": Maintaining Language and Reframing Identity through the Media, 369 Colleen Cotter 20 Constructing Opposition within Girls' Games, 388 Marjorie Harness Goodwin

Name Index, 411 Subject Index, 417

xi

This page intentionally left blank

CONTRIBUTORS

Rusty Barrett Department of Linguistics, University of Texas, Austin Mary Bucholtz Department of English, Texas A&M University Lisa Capps Department of Psychology, University of California, Berkeley Jennifer Coates Department of English, Roehampton Institute, London Colleen Cotter Department of Linguistics, Georgetown University Rebecca J. Dobkins Department of Sociology and Anthropology, Willamette University Marjorie Harness Goodwin Department of Anthropology, University of California, Los Angeles Caitlin Mines Department of English, San Francisco State University William Leap Department of Anthropology, American University A. C. Liang Department of Linguistics, University of California, Berkeley Anna Li via Department of French, University of Illinois, UrbanaChampaign

Norma Mendo/a-Denton Department of Anthropology, University of Arizona Marcyliena Morgan Department of Anthropology, University of California, Los Angeles Graduate School of Education, Harvard University Marjorie Faulstich Orellana School of Education and Social Policy, 3Northwestern University Patricia E. Sawin Department of Anthropology, University of North Carolina, Chapel Hill Laurel A. Sutton Department of Linguistics, University of California, Berkeley Deborah Tannen Department of Linguistics, Georgetown University Sara Trechter Department of English, California State University, Chico Keith Walters Department of Linguistics, University of Texas, Austin Kathleen M. Wood Department of English, Gallaudet University

xiii

This page intentionally left blank

REINVENTING IDENTITIES

This page intentionally left blank

MARY BUCHOLTZ

Bad Examples Transgression and Progress in Language and Gender Studies

Reinventing language and gender scholarship For much of its existence, language and gender research has had an uneasy relationship with feminist theory. Following an initial flurry of interest in linguistic approaches at the beginning of the second wave of feminism in the late 1960s and early 1970s, feminist theory retained its concern with language but soon found less inspiration in linguistics than in surrounding disciplines such as philosophy and literary criticism. At the material level, this tension has been manifest in the disproportionately low representation of linguists, broadly defined, in women's studies programs, conferences, and journals. At the intellectual level, the divide between the two enterprises has been evident until very recently in each field's relative lack of influence on the development of the other. The reasons for this absence of communication are complex. In part it can be attributed to the scientific urge of much linguistic research, a paradigm that has been called into question by numerous feminist scholars (Harding 1991; Longino 1990). Moreover, working within the discipline of linguistics, which especially in the United States has been much slower than the other social sciences to shift its focus from the "science" to the "social" aspect of its intellectual mandate, feminist linguists have been cautious about the use of politically progressive theory in our scholarship. Some of the problem can also be ascribed to the small size of linguistics as a discipline and the resultant difficulty of forming a critical mass of influential feminist work, as our counterparts in sociology, sociocultural anthropology, and literary studies are more easily able to do. For their part, feminist theorists have often framed their concerns in such broad terms that language and gender scholars, with our attention to the fine details of language, have had difficulty seeing ourselves in contemporary discussions. 3

4

R E I N V E N T I N G IDENTITIES

But recent changes in the shape of both fields augur the possibility of productive new interactions between language and gender on the one hand and feminist theory on the other. The rapid expansion of the subfield of discourse analysis within linguistics and the extension of its theoretical and methodological insights to many other disciplines have raised the profile of linguistics within the academy and have provided the conditions in which other branches of linguistics can finally get a new hearing within feminist scholarship. Most crucially, the fundamental observation of discourse analysis, that speakers' identities emerge from discourse, is highly compatible with the social-constructivist bent of much current feminist research. Conversely, feminist theory's vital interest in identity, which is now viewed as a local production rather than an enduring category, addresses a central yet largely uncxamined dimension of language and gender research. With the advent of discursive approaches to identity and with the more contextual analyses of both language and identity that they enable, it may now be possible for language and gender scholars and feminist theorists to pool our resources in a shared intellectual enterprise. Thus despite the current boom in language and gender scholarship, this volume is both timely and necessary: timely because it builds on existing work and necessary because, although its methods derive from linguistics, its theoretical question is taken from feminist theory.' To reinvent identity within language and gender research is to do nothing less than reinvent the field itself. In making the first steps toward a new alliance with feminist theory, the authors in this volume often exceed the boundaries of what is considered the core of language and gender scholarship. By investigating new sites by means of new methods and theories, as well as by casting a new eye on traditional research problems, the contributors to Reinventing Identities transgress—and thereby transcend—the limitations of business as usual both within linguistics as a whole and in the small corner of the discipline inhabited by language and gender. Language and the identity crisis in feminist theory If language and gender scholars are to explore the question of identity, we must enter into dialogue with feminist theorists, for the problem of contemporary feminism is the question of identity. Earlier varieties of feminism either finessed the question (by assuming the universality of the experience of white Western heterosexual women of the middle class) or reduced it to an ontological first principle (by taking an essential difference between women and men as axiomatic). More recent scholarship, however, has engaged the issue fully, recognizing that gender identity is at once more specific than most 1970s feminism realized and more fluid than much 1980s feminism allowed. Debates continue but on different terms: Where previously to dissent from the predominant feminist position was to assert that identity was, in fact, a problem (see hooks 1981; Hull, Scott, & Smith 1982), currently the disagreement lies in how to characterize the kind of problem that identity is (cf. Weir 1996). Indeed, as illustrated by two of the most important books in feminist theory in the past decade, Gloria Anzaldua's edited collection Making Face, Making Soul / Hacienda Caras and Judith Butler's Gender Trouble, both published in 1990, despite considerable

Bad Examples

5

theoretical differences most contemporary feminist scholars agree that identity is far less static than previously thought: "Making faces" is my metaphor for constructing one's identity.... In our self-reflectivity and in our active participation with the issues that confront us, whether it be through writing, front-line activism, or individual self-development, we are also uncovering the inter-faces, the very spaces and places where our multiple-surfaced, colored, racially gendered bodies intersect and interconnect. (Anzaldua 1990:xvi) Just as bodily surfaces are enacted as the natural, so these surfaces can become the site of a dissonant and denaturalized performance that reveals the performative status of the natural itself. . . . As the effects of a subtle and politically enforced performativity, gender is an "act," as it were, that is open to splittings, self-parody, self-criticism, and those hyperbolic exhibitions of "the natural" that, in their very exaggeration, reveal its fundamentally phantasmatic status. (Butler 1990:146-147)

Both Anzaldua and Butler view identity as a construct. In this regard their theories mesh with the insights of recent language and gender scholarship. They differ, however, in the role they assign to the body. For Anzaldua, gender and other aspects of identity are inextricably interconnected, so that bodies are not simply racial and gendered but "racial/y gendered." (See also Morgan, chapter 1, and Orellana, chapter 3, this volume.) Butler, on the other hand, views the body as the stage on which gender is performed, where elements of the self, rather than being uncovered, as Anzaldua proposes, are projected and made to seem natural and, at times, unnatural. Such debates over the body have historically been remote from the concerns of language and gender researchers, but we would do well to attend to the discussions that have developed from the two positions staked out by Anzaldua and Butler. This is so not only because feminist linguists are becoming aware of the importance of the body in language-based studies (see, for example, Goodwin, chapter 20, and Walters, chapter 10, this volume) but also because both theories, like most current strands of feminism, are predicated, as we will see, on the central! ty of language. However, although bodies have come to matter in feminist theory—not only in the broad theoretical sense envisioned by French feminists such as Luce Irigaray (1985a, b) but also in the details of color, age, and so on—feminism has not yet accorded the same attention to the details of language. Hence feminist theory's current interest in the embodiment of language requires a fuller consideration of language itself, a task that language and gender researchers are well positioned to carry out. To avoid replaying the backlash against linguistics in the past, however, it is necessary to understand the history of linguistic analysis within feminism. The linguistic focus of much recent feminist theory has its roots in feminism's second wave. At that time, scholarship on the language used by, to, and about women was part of the feminist project of documenting the dailiness of gender oppression (see discussion in Thorne, Kramarae, & Henley 1983). Most influential were studies of how language itself was implicated in women's subjugation through pejorative referring terms (Lakoff 1975; Schulz 1975; Stanley 1977) and overgeneralized male forms such as the use of the masculine pronoun as generic (Bodine 1975; Dubois &

6

REINVENTING IDENTITIES

Crouch 1987), as well as through power imbalances in cross-sex interactions (Fishman 1983; West & Zimmerman 1983). In the picture painted by such research, all women were oppressed by the overriding force of language. By making language visible and subjecting it to political analysis, language and gender scholarship furthered the feminist aim of uncovering the workings of male dominance everywhere in society. Though belittled or ignored by the scholars' own nonfeminist colleagues, this work had a profound impact on feminists in all disciplines. However, opposition to research within this paradigm soon emerged within feminist circles as well. In the face of critiques, especially by women of color, feminist theory has moved away from an exclusive focus on the undifferentiated category of 'woman' and toward a recognition of the diversity of identities that such a vast cover term obscures. Formerly central issues of language have become less relevant in this context, but the new analytical frameworks pose new linguistic questions. Previously, feminist researchers, both within and outside the field of language and gender, framed the problem in terms of how language affects gender. The question now becomes how language effects gender. Thus where women were once cast as victims of the masculine power of language, they are now viewed instead as active language users in their own right. Yet the issue of power has not been set aside; rather, instead of invoking an invisible and omnipotent patriarchy as the source of power whereby women are interrupted and silenced, researchers have begun to examine the disruptions of this hegemonic system, the moments when women's voices interrupt the dominant discourse and subversive identities break through (cf. Dobkins, chapter 9, and Sutton, chapter 8, this volume). In acknowledging the complexity of women's identities, then, current feminist theorists also acknowledge the complexity of women's relationship to language. During the period of rapid change wrought by feminists of color, the specificity of women's bodies became increasingly important in feminist theory. Likewise, as queer theory developed, the female body as the subject of feminism itself was called into doubt, and the field of gender theory emerged. This complexity is evident in the work of both Anzaldua and Butler. The subjects of Anzaldua's variety of feminism are women of color, who face both sexism and racism as well as other forms of power. The multiple identity positions they occupy offer multiple voices with which to speak: "When we come into possession of a voice, we sometimes have to choose with which voice (the voice of the dyke, the Chicana, the professor, the master), in which voice (first person, third, vernacular, formal) or in which language (Black English, TexMex, Spanish, academese) to speak and write in" (Anzaldua 1990:xxiii). In keeping with the theoretical emphasis of much multicultural feminism, Anzaldua's view of language highlights the multifaceted nature of the self. No aspect of identity has priority and no language linked to these identities is privileged over any other. In fact, by legitimating such "impure" linguistic varieties as "Tex-Mex" (that is, Spanish-English codeswitching; see also Anzaldua 1987) on par with prestigious forms such as "academese," Anzaldua implicitly challenges the cultural hierarchy in which certain languages and the identities that they render are valued over others. Choosing such a voice is an act of rebellion against the power of linguistic ideology. She thus clearly separates herself from earlier feminist preoccupations: agency, choice, and voice here replace passivity, oppression, and silence.

Bad Examples

1

Butler, too, considers how language mediates between the individual and wider cultural hegemonies: "Language is not an exterior medium or instrument into which I pour a self and from which I glean a reflection of that self.. . . Indeed to understand identity as ^practice, and as a signifying practice, is to understand culturally intelligible subjects as the resulting effects of a rule-bound discourse that inserts itself in the pervasive and mundane signifying acts of linguistic life" (Butler 1990:143-145). Butler suggests that selfhood is manufactured through language. Identity is not a category at all for her; instead it is a semiotic activity whereby individuals are made to make cultural "sense." Yet there is space for dissent in this framework: Those who resist the dictates of the culture by troubling its categories highlight the constructed nature of these divisions. 2 In this volume, some scholars implicitly align themselves with Anzaldiia and note the multiplicity of selves available to speakers (see, e.g., Barrett, chapter 16; Coates, chapter 6; Cotter, chapter 19; and Orellana, chapter 3), as well as the multiplicity of identities within what is often wrongly seen as a monolithic social category (Goodwin, chapter 20; Mendoza-Denton, chapter 14; Morgan, chapter 1; Sawin, chapter 12). Other researchers focus, as Butler (1990) does, on the hegemonic linguistic forces against which women and girls struggle and to which they sometimes capitulate (Capps, chapter4; Dobkins, chapter 9; Mines, chapter 7; Livia, chapter 17) and on the linguistic practices, filtered through cultural ideologies, through which gender identities, both "culturally intelligible" and otherwise, are produced (Leap, chapter 13; Tannen, chapter 1 1 ; Trechter, chapter 5; Walters, chapter 10). Still others bring the two paradigms together by underscoring the interaction between fluid identities and rigid social structures (Bucholtz, chapter 18; Liang, chapter 15; Sutton, chapter 8; Wood, chapter 2). All these authors take intellectual risks—methodological, theoretical, and political—to advance the field of language and gender. In so doing they follow a time-honored tradition in language and gender studies of challenging the boundaries of permissible, respectable research. The past decades of scholarship provide numerous such "bad examples" for our contemplation: daring studies of the relationship between language and gender that were often criticized or misunderstood in their own day or that, though set aside too soon, have made crucial contributions to the present state of the field. 3 Like much of this previous work, the present studies are especially notable for their use of feminist and other social theory and their farsighted analyses of identity. Also significant is their recognition of identity as a practice rather than a category, an actively constructed performance rather than a preexisting role; their concern with identities that deviate from the traditional focus of scholarship; and their consideration of the balance between individual agency and structural inequality. Bad habits: Identities in practice and performance Much language and gender research in the past two decades has subscribed to the social-scientific premise that identity is a category that individuals inhabit. Analyses compared the linguistic behavior of women and men (usually in cross-sex interaction) as aggregates and did not discuss within-group differences. Membership in such

8

REINVENTING IDENTITIES

groups, as well as in lesser-discussed groups based on race, ethnicity, or class, was viewed as self-evident. Meanwhile, locally defined groupings based on ongoing activities and concerns were rarely given scholarly attention; if they were, members were assigned to large-scale categories of gender, race, ethnicity, and class. During the same period, feminist theory was undergoing the above-described transformation initiated by the work of women of color, and it abandoned the category model of identity. Consequently, language and gender studies, which had in the 1970s been central to the feminist project, separated from other feminist research. Not surprisingly, then, Penelope Eckert and Sally McConnell-Ginet's (1992) introduction of the community-of-practice model quickly revolutionized language and gender research. Although formulated in terms of community, the theory is easily applied to identity as well, for it shifts the focus from groups as undifferentiated categories to complex configurations of individuals acting in part together and in part separately. This new focus on individuals is part of a wider urge in sociolinguistics and language and gender (Johnstone 1996; Johnstone & Bean 1997). In this volume, for example, Keith Walters (chapter 10) revisits cross-sex interaction in a way that highlights the situated, embodied experiences of the individual speakers, and Norma Mendoza-Denton (chapter 14) and Patricia Sawin (chapter 12) each, in different ways, demonstrate how individual and cultural specificities interact in language use. Eckert and McConnell-Ginet's emphasis on action as the basis of community renewed an interest in local, activity-based identities.4 Participants in such activities assume a variety of roles and identities whose relevance varies from moment to moment. In this volume Colleen Cotter (chapter 19) and Deborah Tannen (chapter 11) both show how in workplace communities of practice, an understudied arena in language and gender studies compared to private contexts, "the difference that difference makes" is specific to speaker, situation, and other factors. The display of the self in public settings also suggests another practice-based aspect of identity: performance. Widely used (though with a narrower definition) in linguistic anthropology (e.g., Bauman 1977;Briggs 1988;Hymes 1981) before it came to be popularized in gender theory (Parker & Sedgwick 1995), the concept of performance in current language and gender research highlights the fact that membership in particular communities and the identities they authorize is achieved rather than assigned: As Kathleen Wood (chapter 2) demonstrates, coherent lesbian or Deaf identities arise through participation in community practices that stand partly in opposition to hegemonic ideologies of heterosexuality and the hearing world. Such violation of biased cultural norms may lead to the misclassification of speakers as deviant or deficient: Marjorie Goodwin (chapter 20) argues that when girls play with the rules rather than by the rules in hopscotch games, researchers may fail to see that girls' games have any rules at all, when in fact girls' creative adaptation of preexisting structures allows them to project a range of local interactional identities. A practice-based analysis thus forces scholars to reassess apparent inadequacies of speakers; one of the great strengths of the community-of-practice model, especially as originally articulated by education theorists Jean Lave and Etienne Wenger (1991), is that "marginal" members—novices, learners, and so forth—move to the center of analysis. In this dynamic framework, practice takes on a double meaning, suggesting both habitual social action and rehearsal for later social action—that is,

Bad Examples

9

for performance. Jennifer Coates (chapter 6) and William Leap (chapter 13) both invoke the second sense of practice in their discussions of linguistic, gender, and sexual socialization of adolescents. For Coates, adolescent girls' friendship talk provides a "backstage" (in Goffman's 1959 sense) in which to "try on" identities; Leap shows how gay adolescents start their "gender careers" by exploring gay resources in books, films, and other media. Perhaps most important, the focus on practice in recent work reminds linguists that language is only one social activity among many and that it takes much of its meaning (both social and referential) from the other practices that surround it. Many contributors to this volume, recognizing the importance of such contextualizing information, provide images and rich descriptions of the words and activities of language users. Attention to such practices is often dismissed as "not linguistics," but like Goodwin's hopscotch players, the contributors to Reinventing Identities insist that breaking the rules is part of the game.

Bad girls: Transgressive identities As the new focus on the margins of community membership may imply, another way that innovative language and gender researchers have set a "bad example" in their research is by using "bad examples" of speakers and linguistic phenomena. Much of the scholarship in language and gender has been what might be called "good-girl research": studies of "good" (that is, normatively female—white, straight, middleclass) women being "good" (that is, normatively feminine). Such research has been a necessary starting point. Yet this definition of what counts as a "good" example excludes many groups and practices. In fact, as Marjorie Orellana's research suggests, how girls use language to experiment with "badness" is itself a crucial question for feminist linguists. Being "good" is not a natural attribute but one constructed through the interplay of language and social expectation. "Bad" girls and women may pose problems for neat theories and hence be eliminated from research as atypical. What research does exist often succumbs to the urge to pathologize or exoticize such speakers. Conversely, speakers who conform precisely to cultural stereotypes of femininity, such as the drag queens in Rusty Barrett's study (chapter 16), are nevertheless almost entirely overlooked in language and gender research, not because of their anomalous femininity but because of their anomalous femaleness (but for examples of the growing scholarship on language and gender transgression, see Gaudio 1997; Hall & O'Donovan 1996; Kulick 1996). Barrett's chapter demonstrates the limitations of traditional definitions of the speech community as the locus of language and identity, for as developed in sociolinguistics by William Labov (1972a, b), membership in the speech community is a measure of cultural and linguistic authenticity as determined by gender, class, ethnicity, and other social factors. The artifice of many linguistic performances and practices cannot be accounted for in the Labovian framework. My study (chapter 18) of the shopping channel as an artificial community that constructs its own authenticity from consumers' linguistic practices indicates that authenticity is itself a production, not an objective measure of community membership and identity.

10

R E I N V E N T I N G IDENTITIES

To display a transgressive identity is to risk not only exclusion but also retribution. Thus as A. C. Liang (chapter 15) shows, gay and lesbian speakers whose identities may put them in danger develop self-protective linguistic practices that allow them to reveal their identities only to those who are likely to be sympathetic. In a world where simply being can count as being bad, identities are often constructed in opposition to dominant cultural ideologies.

Bad subjects: The politics of gender identity Such resistant identities recall Louis Althusser's famous declaration regarding the "bad subjects" of the state: "the 'bad subjects' . . . on occasion provoke the intervention of one of the detachments of the (repressive) State apparatus. But the vast majority of (good) subjects work all right 'all by themselves,' i.e. by ideology . . ." (1971:181). In the battle between the bad subjects and the forces of hegemony, language is a doubleedged sword. Numerous social theorists have recognized the close relation between language and systems of power (see, e.g., Bourdieu 1991; Foucault 1980). Caitlin Hines (chapter 7) shows how this relationship plays out in the English lexicon, and Sara Trechter (chapter 5) reveals its workings in the verbal system of the Lakhota language. These analyses of gender ideology improve on earlier work: Hines demonstrates both the details of sexism in the linguistic system and the details of the linguistic system in sexism, and Trechter shows how the system is used in practice. Practice-based research must also recognize the dynamism of language and thus the possibility of changing structural biases. Anna Livia (chapter 17) recounts the history of one widely debated attempt at feminist linguistic engineering. Women's agency, as exemplified by Livia, is in fact a central component of new research on the power of cultural ideologies in women's lives. Whereas earlier work portrayed women as passive victims of male power and privilege, recent scholarship also recognizes women's active contributions even in situations of extreme power imbalance. At times, women's struggle for autonomy and self-definition in interaction with men may only result in greater disempowerment, as Lisa Capps (chapter 4) shows. At other times, women's efforts are more successful, although they may be forced to rely on the linguistic tools of their oppressors to insist on their own identities— Rebecca Dobkins's study (chapter 9) of Native American women's letters to federal officials offers an illustration of this situation. Yet women may find alternative discursive forms, even, as Laurel Sutton describes in her work (chapter 8) on underground zincs and online publications, reshaping male-dominated discourse domains to accommodate their expressions of resistant selves. As women change the shape of cultural discourse, the contours of intellectual discourse are changing as well. Earlier critiques concerning the homogeneity of feminist theory have also had effects in language and gender scholarship. Thus the present volume features work on Latinas by Goodwin, Mcndoza-Denton, and Orellana; on African Americans by Barrett and Morgan; on Native Americans by Dobkins and Trechter; and on non-Western women by Walters, as well as on lesbians and gay men by Barrett, Leap, Liang, and Wood. Many chapters in this book also contribute implicitly or explicitly to the recent project to interrogate previously invisible hcge-

Bad Examples

11

monic categories such as whiteness, masculinity, heterosexuality, and the middle class (see the chapters by Barrett, Bucholtz, Wood, and others). New approaches to power in language and gender scholarship likewise require that linguists recognize their own complicity in the reproduction of inequities based on gender, race, class, and other factors. Marcyliena Morgan (chapter 1) describes the social and cultural censure that many African Americans level against the act of "breaking bad," or pretending to a level of knowledge that one does not in fact have. Unlike the other forms of researcher "badness" delineated above, "breaking bad" cannot be recuperated for feminist use, for it results in acts of scholarly negligence and harm ranging from wrongful omission to misrepresentation to out-and-out inaccuracies. Both Morgan and Sara Trechter point out instances in which linguists represent dominated social groups as exotic "others" who are portrayed as fundamentally different from the normative, dominant group. Their work reminds language and gender scholars that reflexivity must always be a part of research and that as we move further from the confines of "goodgirl" research we have an increasingly heavy obligation to describe speakers and the worlds in which they move without the distorting effects of our own cultural and intellectual ideologies about how such speakers "should" or "must" be. Considering how linguistics itself has contributed to the promotion of such ideologies will help ensure that as we strive to be "bad examples," we do not produce bad research.

Overview of the volume Taking seriously the diverse resources available to language users for identity construction, the chapters of Reinventing Identities draw on a variety of linguistic forms and contexts. Thus the authors examine every linguistic level, from phonetics and phonology to discourse phenomena, from lexical and semantic issues to broader pragmatic processes. In the realm of discourse, contributors turn to conversation, narrative, and activity-based language use, as well as to written language and media representations of the social world, recognizing that the practices associated with all such linguistic arenas may be conscious attempts to subvert social norms or the result of the speaker's effort to synthesize public displays and private imaginings of the self. The authors also consider a range of identities, from the most locally and interactionally grounded participant roles to the broadest transhistorical stereotypes, from identities rooted in race, ethnicity, sexuality, and class to those based on membership in social groups and communities of practice to those associated with largescale social categories and imagined communities produced by industrialization, technology, and nationalism. This recognition of the multiple manifestations of the self in linguistic practice allows contributors to Reinventing Identities to avoid the narrowness of earlier analyses and to explore dimensions of identity hitherto uninvestigated in the field of language and gender. The volume is divided into four parts, which describe the ways that identities, both chosen and imposed, are created, shaped, and altered across times, places, speakers, and contexts. The first part, "Identity as Invention," locates the originary moment in the construction of a variety of gendered identities. Next, "Identity as Ideology" explores the manifold ways that larger social forces may impose and perpetuate

12

R E I N V E N T I N G IDENTITIES

identities through language use. "Identity as Ingenuity" addresses the creative strategies that speakers employ in response to hegemonic expectations. Finally, "Identity as Improvisation" examines the emergence of new identities that rework traditional formulations of selfhood while engaging language in innovative ways. Identity as invention The chapters in part I make visible the often invisible process of identity construction through language use, thereby challenging the essentialist notion that gender identities are inevitable, natural, and fixed. All the chapters share the insight that identities, far from being given in advance for individuals to step into, emerge over time through discursive and other social practices. Nor is identity construction an exclusively individual act; instead, social selves are produced in interaction, through processes of contestation and collaboration. The section opens with Marcyliena Morgan's overview of research on African American women's speech. African American linguistic practice holds a central place in the consolidation of sociolinguistics as a discipline, and Morgan points out that this situation has led to the generalization of models of African American identity that were first developed in research on lower-class vernacular-speaking urban adolescent male gang members. The speech of those who fall outside these parameters— especially female and middle-class African Americans—has been marginalized in research on the grounds that it is less authentic. Morgan corrects the biases of prior research by focusing on the language use of African American girls and women at different life stages. She shows that speech events such as signifying, instigating, and reading dialect have different degrees of social salience for children, teenagers, and adults, each of whom may enact them in different ways and for different purposes. These speech events are united, however, by their role in the interactive construction of "face" as part of a speaker's social identity and as part of a specifically African American cultural identity. Morgan's attention to the temporal dimension highlights the fact that identities are not static or universal across members of a group but are shaped over time through dialogic processes among speakers. The creation of a shared cultural identity is also central to the work of Kathleen Wood (chapter 2) on Deaf and hearing lesbians' tellings of coming-out stories on the Internet, in writing, and in American Sign Language. Contrary to theories of lesbian identity that posit a priori commonalities among lesbians, Wood argues that the lesbians in her study actively create their on-line community through shared linguistic practices. Although such stories vary widely in their structure and content, they have in common their engagement with ideologies of normative heterosexual!ty, to which they respond. Wood's findings thus force a revision of frameworks that assert that local communities are structurally self-contained entities. Additionally, her research suggests that lesbian identity may intersect with other dimensions of the self such as Deafness. Like Morgan's findings for African American speakers, Wood's study indicates that culturally recognized speech events may be used to construct social identities. Moreover, the identities produced in coming-out stories may vary according to the resources available in the communicative medium and are therefore closely bound to the circumstances of narrative production.

Bad Examples

13

Wood's insight that narrative content and communicative medium work together to create identity is supported by Marjorie Faulstich Orellana's research (chapter 3) on the written narratives of Latina and Latino children from low-income families in a Los Angeles classroom. Orellana examined the in-class production of studentauthored books over the course of many months and found that children often wrote themselves into their stories as a way of trying on possible identities. Many of these fictive identities—such as superheroes, racing champions, and so on—projected students into positions that challenged the constraints of racism and poverty, but most storylines reproduced gendered expectations about who and what students could be. Girls often wrote stories that emphasized the importance of being good, whereas naughtiness was much more prominent in boys' stories about themselves. Such ideological restrictions were even more pronounced when students cast one another in their stories—for example, boys often placed girls in supportive or passive roles. Orellana concludes that students' imaginative writing is closely tied to the realities and possibilities of their lives, and although free-writing classes like the one she studied may have some liberatory potential, for many students, especially working-class girls of color, storytelling is an activity fraught with issues of power and inequality. The recognition that identities may be simultaneously chosen and imposed through language use, a recognition that is fundamental to Orellana's research, is also central to Lisa Capps's work (chapter 4). Likewise, both authors demonstrate that seemingly cognitive or psychological phenomena are deeply social in their origins and effects. Whereas Orellana's work complicates cognitive theories of literacy, the tendency for theorists to root identities in naturalized and asocial explanations such as biology or psychology is dramatically challenged by Capps's research. Capps examines the interactional construction of agoraphobia, a condition in which an individual fears and avoids places and situations she perceives as dangerous. Many researchers who study this syndrome, which disproportionately affects women, attribute it solely to the sufferer's psychological makeup, but Capps demonstrates that women who have been diagnosed as agoraphobic are also socially constructed as irrational through interaction with family members. Based on a long-term case study of an agoraphobic woman and her family, Capps shows that the ordinary narratives that family members tell at the dinner table are important sites for the discursive linking of agoraphobia and irrationality. The responses of the woman's husband, as recipient of her stories, are particularly crucial in this process, which unfolds over a series of interactions. Yet the woman herself is not merely a passive victim of her husband; indeed, Capps traces a complex pattern in which the woman asserts her rationality through specific discourse strategies, but these very strategies trigger responses from her husband that insist on her irrationality; thus mental illness is here revealed to be a joint construction that has much of its source in family dynamics. Like Morgan and Wood, Capps identifies the temporal component of this pattern, but here discourse is a resource not for shared identities but for the distancing and differentiation of self and other. Social differentiation is also at issue in Sara Trechter's investigation (chapter 5) of the creation of "women's language" and "men's language" as distinct categories in the Native American language Lakhota. Trechter's focus on language ideology at the cultural level complements Capps's analysis of how identities come to be im-

14

R E I N V E N T I N G IDENTITIES

posed in the most intimate setting, that of the family. Like Capps, Trechter shows that a category often thought of as natural is instead historically contingent. Although some linguists present "women's language" and "men's language" as direct and exclusive expressions of gender identity, they are in fact cultural products that index a variety of social and interactional meanings. Trechter also makes clear through an examination of gendered final particles in the Lakhota verb system that although speakers may choose not to use the "appropriate" gendered form in a particular context, the cultural ideology that maps linguistic difference to gender difference is always available for observers to render anomalous speakers "culturally intelligible." Through observers' metalinguistic commentaries, interactional resources gradually come to be linked to gender. Identity as ideology The authors writing in part II take as their common theme the role of ideology in the construction of identity—not merely how speakers conform to an accepted or imposed ideology, but how they rebel against or subvert a powerful system of beliefs. As the authors show, ideological systems themselves exist as cultural constructs, subject to processes of change and revision by individuals and groups. Identity thus acknowledges ideology as one of the many elements in its framework. Jennifer Coates (chapter 6) examines the ways in which gender beliefs and practices are manifested in the conversations of teenage girls. The years between childhood and adulthood are crucial in the individual's exploration of identity; it is here that society's messages and pressures may come to dominate a young woman's sense of herself. Using data collected over a 3-year period in a friendship group of four British girls, Coates documents the changes in discourse type and topic that reveal the process of gender construction. She shows how the girls' sense of their femininity is at times contradictory and precarious; they experiment with a range of discourse styles and subject positions but become increasingly restricted to more serious genres, avoiding the playful discourse of their early adolescence. Central to their new, adolescent discourse is what Coates terms "consciousness-raising / self-disclosure," a form of support for one another that positions the girls as victims of external forces (including their own bodies). But this period of girls' diminished sense of personal agency is temporary, serving them in later adolescence as a marker of the difficulties of moving from the freedom of girlhood to the constraints of womanhood. Coales shows that one of the greatest pressures facing girls as they grow up is the set of cultural ideologies concerning women. Caillin Mines explores one such ideology in her work (chapter 7) on the "woman as dessert" metaphor that pervades English. This metaphor is not merely a linguistic phenomenon but a larger cultural construct, for it has nonlinguistic reflexes, for example, in visual representations of women. Her research thus provides a much-needed analysis of the gender-based ideology inherent in the construction of an entire range of "woman as object" metaphors and opens the way for further examination of the place of gendered subtexts in long-established linguistic conventions. Mines uses a cognitive framework to establish the implications of women's imposed identity as sweet, possessable objects. Searching through many types and realizations of this image, she provides a thor-

Bad Examples

15

ough deconstruct!on of its history, form, and use. She finds that multiple linguistic levels from phonology to semantics work together to construct a coherent metaphorical system that has been a social resource for the representation of women over hundreds of years and can continue to be extended indefinitely. Her data show how a single ideologically created identity, once fixed in the language, permeates the discourse of speakers far removed from its source. The channels of transmission of cultural ideology are the main focus of Laurel Sutton's work (chapter 8) on the status of popular media in language and gender studies. Contrary to traditional and mainstream linguistic practices of using only "authentic" data—that is, spontaneous spoken discourse—to establish patterns of language use, Sutton calls for the inclusion of all forms of expression in the analytic corpus, a move that has been initiated primarily by researchers of language and gender and other politically motivated scholars. She advocates, in particular, greater attention to often ephemeral forms of individually produced media that rely on new technologies for their production and circulation, such as feminist zincs (selfpublished, low-circulation magazines) and on-line journals. Given the extent to which the mass media shape identity, especially for young women (such as those studied by Coates), the existence of such alternative media offers the possibility of forming an identity that is consciously resistant to prevailing cultural expectations. However, as Sutton notes, the publicizing of private selves in such media also opens their authors up to attack from apologists for the dominant culture. Such media—and their representations of alternative selves—therefore disappear quickly unless researchers make the effort to rescue them for study. Another example of the recovery of dissenting voices comes from Rebecca Dobkins (chapter 9), who looks at letters written between Native American mothers and European American boarding-school officials in the early twentieth century. Letters were the only form of communication permitted these parents in their struggle to retain control over their children's place in their families and their children's larger identities as Native Americans. From Indian women's one-down position as racial and cultural minorities in a white society, identity was not negotiable, for Indian culture faced potential eradication daily at the hands of governmental policies of assimilation and cultural destruction. The Indian mothers' acts of resistance, made manifest in their letters, reveal the structure and mechanics of the official school authority, as well as the ruptures within it. The women who wrote these letters variously manage and challenge the language of the dominant culture. They thus recognize their roles both inside (as good American parents who, in their children's interest, cede their power to the government) and outside (as good Native American parents who reassert their cultural ties by bringing their children home) the governmentimposed ideology of the time. Finally, Keith Walters (chapter 10) looks at a contemporary case of gender and culture negotiation in Tunisia. Walters, like Hines, examines how the words used by speakers reflect cultural assumptions about identity and the roles women and men must fill. Here, the added dimension of religious ideology and the complexity of language choice raise important questions about the difference between imagined and practiced beliefs about gender. Walters considers the evidence of a sociolinguistic interview conducted between a female interviewer and a male interviewee. The two

16

R E I N V E N T I N G IDENTITIES

speakers shared a common cultural background and were of the same age, but the interviewer was of a higher educational and social status than the man she interviewed. Their interaction was already anomalous by the standards of this small-town Tunisian Muslim community of the 1980s, because free conversational exchange between unrelated women and men is not usually permitted. The speakers used linguistic resources to stake claims about their own identities and ideologies of gender and equality, which quickly came into conflict and were symbolized through the use of codeswitching and variant pronunciations of the crucial word 'woman'. Walters shows that in this encounter the two speakers use language to create their own embodiment as gendered, classed, and otherwise categorized individuals. Their struggle thus emblemizes the disjunction between ideologies of identity imposed by both researchers and cultural members and the way disparate elements of the self play out in interaction at moments of cultural shift. Identity as ingenuity Whereas the second part of Reinventing Identities treats the role of ideological forces in the construction of identity, part III locates identity at the intersection between culturally imposed and personal meanings. The chapters in this section explore how language serves as a vehicle for speakers' creative responses even as their linguistic behavior reflects the stability of existing social categories. Part III begins with Deborah Tannen's analysis of two same-sex hierarchical office interactions. The chapter underscores the point that gender in particular and identity in general arc the outcome of negotiation, by which speakers contest and give way to the social forces that shape and constitute identity. The complexity of the relationship between gender and power is highlighted through the concept of framing, that is, through analysis of the moment-by-moment alignments taken up by participants in interaction. Within this theoretical framework, small talk in work settings is seen to instantiate the negotiations between connection and status among participants. Higher-status participants have the power to frame the interaction, whereas lower-status participants are limited to contesting or acceding to the frame. The ways in which this negotiation takes place are, Tannen argues, borrowing a term from Erving Goffman, "sex-class-linked"—that is, linked to the class of women and the class of men. Whether constituted in gender difference or power imbalance, identity is therefore best understood as a matter of display, emerging through the course of interaction, not intrinsic to a particular individual. By bringing together two widely invoked theoretical approaches to gender and language—cultural difference and dominance—Tannen offers a balanced reconceptualization of gender identity. Tannen articulates the phenomenon of gender as a performed display by which individuals, although constrained by the contingencies of interaction, nevertheless pick and choose from among an inventory of practices. Her implicit critique of essentialist conceptions of social categories is shared by Patricia Sawin (chapter 12), who raises several fundamental methodological and theoretical issues for the study of "women's narrative." Sawin observes that prevailing narrative models fail to account for the structure of the stories told by an elderly Appalachian woman who is the subject of her study. The discrepancy between data and theory prompts her to

Bad Examples

17

rethink the widespread concept of "women's narrative." She notes that this concept is an overextension and reification of theoretical models originally posited for a circumscribed set of social groups, primarily white middle-class women in the United States and England. In light of her findings, she theorizes identity and narrative as the result of a combination of interactional needs and existing social structures. According to Sawin, the study of women's stories of self as a means of identity construction requires more than the analysis of isolated narratives. Instead, narrative is better seen as the implementation of communicative strategies developed both in accordance with and in opposition to dominant social structures and discourses. Tannen and Sawin are primarily concerned with critiquing the theoretical implications of traditional understandings of identity. By contrast, William Leap and the remaining authors in this part examine the creation of identity from the point of view of the agent and how she or he works within and against cultural ideologies of identity in order to construct a consciously chosen identity. In his examination of the gay male coming-out experience in chapter 13, Leap considers American gay identity as the outcome of a struggle to verbalize unnameable feelings. The transition between culturally dictated heterosexual reproductive roles and personally created gay identity, he suggests, takes place via a process of gay socialization akin to language acquisition through which speakers learn culture-specific text-building strategies. Leap's emphasis on gay identity as acquisition and process rather than as static essentialism calls attention to the socially constructed aspects of homosexuality. "Coming into gayness" is thus not a status achieved through biology or by the comingout act alone but rather is an ongoing practice of learning and performing the many dimensions of a shared cultural identity. Like Leap, Norma Mendoza-Denton (chapter 14) is also concerned with the processes by which individuals arrive at their individual and group identities. On the basis of the play-by-play interaction of a group of Latina teenagers, she shows how scholars who approach girls' and women's talk as cooperative miss the intricate strategies that speakers employ to align and disalign with one another at a variety of levels. The Latina speakers use turn-initial no as a discourse marker that signals these ongoing affiliation processes, which are based in part on shared Latina identity, on regional and class background, and on social attitudes and beliefs. At the same time, alignments may be disrupted by speakers' different pragmatic assumptions that may be remedied only by explicit metalinguistic work. Mendoza-Den ton's research suggests that neither interaction-based nor category-based approaches alone are adequate for the understanding of identity in this context; instead, researchers must consider as well speakers' linguistic and pragmatic systems, their social backgrounds, their relationships to one another, and their individual beliefs and opinions. A. C. Liang shares Mendoza-Denton's concern with the role of pragmatics in the production of identity. In chapter 15, she demonstrates how people can exploit the inherent ambiguity in language to create an equivocal identity. Examining the potential of Gricean pragmatics as a framework for the study of gender and sexuality, Liang assesses the adequacy of the Cooperative Principle of conversation against real-world examples, focusing in particular on the relationship between homosexuals and heterosexuals. She argues that H. P. Grice's (1975) notion of cooperativeness is dependent on power relations, and she questions the necessity of his assump-

18

REINVENTING IDENTITIES

tion of common purposes or mutually accepted directions for the exchange of information. Given the risk to their physical and mental well-being if they elect to come out to their interlocutors, lesbians and gays may choose to employ communicative strategies from which their covert meanings, although misleadingly worded for straight listeners, can be properly inferred only if listeners disabuse themselves of the default assumption of heterosexuality. These creative strategies of simultaneous masking and display of sexual identity allow lesbian and gay speakers to protect themselves from the dangers of homophobia while locating themselves in solidarity with listeners who are familiar with these strategies. The ambiguous selves that lesbian and gay speakers project in this way exemplify the limitations of either universal pragmatic or categorical sociolinguistic models of identity. Identity as improvisation The process of identity creation and consolidation is not a one-time event. New social arrangements provide the means to shape new identities. As the chapters in part IV of Reinventing Identities demonstrate, resources as diverse as barroom entertainment, science fiction, mass media technology, and children's games arc employed by speakers and writers in ways that cannot be predicted in advance. Through the innovative reworking of previously formulated structures, language users go beyond deterministic frameworks of identity, arguing through their actions that identity is instead a continuous creative practice. Like Liang's, Rusty Barrett's research (chapter 16) reveals how speakers at once work within and subvert dominant ideologies in their construction of identity. His chapter opens part IV with a critique of the concept of identity in theories of sexuality, gender, and race. Based on an ethnographic study of African American drag queens' performances in a gay bar, Barrett's work argues against privileging a particular aspect of the performers' complex identities. As he demonstrates, race, gender, class, and sexuality become salient at different moments during drag performances. Barrett counters feminist charges that drag queens want to be women and the claims of some African Americans that black gay men want to be white; he shows that African American drag queens use the features of a stereotyped "white middleclass women's language" to challenge racism and homophobia. However, Barrett does not embrace the position of queer theorists, influenced by J. L. Austin's (1962) concept of the performative utterance, that drag is a celebratory performative assertion of the dissolution of gender categories. He notes that misogyny underlies some drag queens' routines, a fact that calls into question the usefulness for radical politics of the performative theory of identity. Anna Livia's study (chapter 17) of nongendered pronoun systems in feminist fiction shares with Barrett's work a concern with how language users play with gender categories for political purposes. Li via notes that epicene (or common-gender) pronoun systems were frequently proposed as part of feminist activist efforts in the 1970s, yet such systems never enjoyed any currency even in feminist communities, although they occur in several futuristic and Utopian feminist novels of the 1970s. Livia argues that epicene systems are meaningful only in juxtaposition to genderbased systems, and she shows how authors use both to contrast their Utopias to sexist

Bad Examples

19

realities. She traces the development of several strategies that authors use to decouple gender and reference, including generic feminine, generic masculine, and invented nongendered forms. Li via's research contextualizes and recovers a moment in radical feminist linguistic history that is too often dismissed as mere "Whorfianism." Her study enjoins practitioners of text linguistics to take more seriously the political effects of fiction. The fictional worlds explored in Livia' s research find their counterpart in the virtual communities of television that are the subject of chapter 18. In this chapter I examine a relatively uninvestigated arena of media discourse: the shopping channel, a network that combines elements of advertising, talk-show discourse, and private conversation. I argue that the shopping channel, which targets primarily lower-middleclass women, uses a variety of visual and discursive strategies to create a fictive community in which middle-class identities are displayed by telehosts and linked to the purchase of the network's products. Although viewers may jointly construct the shopping-channel community and their identities within it through on-the-air conversations with the network host, the prevailing discourse, which furthers the channel' s capitalist agenda, rests on gender and class ideologies. Callers' nonstandard language becomes another commodity that hosts use to construct their own positions of authority and middle-class identity and to position callers as authentic and enthusiastic endorsers of the network's products. Yet the power relations thus instantiated are destabilized by the demands of capitalism, so that viewers also derive a measure of discursive and economic power from the shopping channel, though always on the terms of the network itself. Colleen Cotter (chapter 19) likewise investigates the complex power dynamics of media discourse, demonstrating in her study of Irish-language radio that the media can be a positive cultural and linguistic resource rather than a global homogenizing force. In so doing, she challenges the viewpoint held by many linguists that the media are a key component in the demise of endangered languages. After describing how the media are employed as a way of promoting Irish-language use and national identity in a late-twentieth-century discourse domain, Cotter examines the place of gender in Ireland's language revival effort at both the microlinguistic and the macrolinguistic levels. She observes that Irish-language use on the air is not dramatically different for women, who have a central place in all aspects of production at the radio station she studied. In fact, in an interaction between a female interviewer and interviewee, "gendered discourse" does not surface at all, for both participants are fulfilling nongendered roles appropriate to the media setting. This production of communicative competence in the media realm promotes competence in Irish as well. The result for women, who have traditionally been at the forefront of language retention efforts worldwide but in private roles, is that they occupy more publicly visible positions within the historically circumscribed contexts of media, language preservation, and gender. Marjorie Harness Goodwin (chapter 20) closes the book with a discussion of how identities arc shaped not through language alone but through entire activity systems. Goodwin's research, like that in all the chapters in this section, demonstrates that established discursive conventions may be employed in unexpected ways and with unanticipated effects. Her investigation of Latina girls' interactional practices

20

REINVENTING IDENTITIES

while playing hopscotch revises earlier understandings of gender and ethnicity. As she points out, girls' games have often been theorized as cooperative activities with undeveloped rule systems that promote loose and nonhierarchical social structures. Moreover, Latinas are frequently stereotyped as noncompetitive and submissive. But Goodwin finds that the Latina girls she studied are fully capable of competitive and self-promoting behavior which, far from deviating from a normatively cooperative game structure, is built into the very fabric of play. Attempts to cheat and resultant efforts to identify and challenge rule violators pervade the hopscotch games Goodwin observed. Her detailed examination of how identities are linguistically built within situated activities offers an alternative framework for understanding gender and other dimensions of identity. The embodied activity that Goodwin considers introduces space, as well as time, to the study of linguistic practice, and the purposeful temporary identities that are constructed and reconstructed from moment to moment in social action are shown to have significance for speakers' more lasting identities based on age and friendship networks. Goodwin's chapter thus brings together the central themes of Reinventing Identities: that language users' identities are not essential to their natures but are produced through contingent social interactions; that these identities are inflected by ideologies of gender and other social constructs; that speakers, writers, and signers respond to these ideologies through practices that sometimes challenge and sometimes reproduce dominant beliefs; and that as new social resources become available, language users enact and produce new identities, themselves temporary and historical, that assign new meanings to gender.

Conclusion If the 1970s were the decade of discovery in language and gender research and the 1980s were a period of critique and correction of earlier work, the 1990s have turned out to be a transformative decade in which researchers have both revisited familiar territory with new tools and set forth into unexplored areas. We are in a period of "rethinking" the basic precepts and concepts of language and gender research (Bergvall, Bing, & Freed 1996; Taylor 1997). Like feminism itself, now undergoing a third wave of scholarship and activism, the field of language and gender has become an increasingly recognized academic enterprise while still maintaining the clarity of perspective that comes from being situated at the intersection of traditional disciplinary boundaries. A rapprochement between feminist linguistics and feminist theory now seems more possible than at any time since the 1970s. Linguistics can offer a theoretically grounded view of language in practice, performance, and ideology, which is vital to present theorizing. Conversely, the interpretive turn in language and gender research and scholars' increasing willingness to look to feminist, as well as linguistic, theory have resulted in a new trend toward theoretically informed scholarship that addresses issues of fundamental concern to feminists in other fields. Most significant in this regard is the fact that the discipline has shifted away from a comparative framework in which discovering differences in the linguistic behavior of women and men as

Bad Examples

21

groups is the central research goal toward an approach rooted in the details of context, concerned with locally meaningful social groupings rather than global gender divisions and attentive to individual variation within, as well as across, gender categories. In her historical discussion of feminism and pronoun politics, Anna Livia reminds those who would trivialize the efforts of the radical-feminist linguistic activists of the 1970s and 1980s that "one had to go there to get here." In other words, we cannot simply denounce research rooted in now-unfashionable theories, for these and other "bad examples" have ushered in entirely new theoretical paradigms. The shape of language and gender scholarship is tied, perhaps more than we sometimes acknowledge, to the shape of feminist theory. Mindful of our discipline's history, then, we must look to feminist theory, past and present, to get our bearings as we move into a new era of research. It is my hope that in the current period of change this volume will contribute to the growing urge toward theory in language and gender scholarship and inspire future researchers to transgress the existing boundaries that separate "good" and "bad" research in the investigation of the linguistic production of identity. NOTES

My thanks to Caitlin Hines, A. C. Liang, Deborah Tannen, and Keith Walters for offering numerous useful suggestions on earlier versions of this introduction and to Jon McCammond for ongoing discussions about the issues explored here. I am also grateful to Kathryn Galyon for her skillful copyediting of this essay. Thanks are due as well to Peter Ohlin and Cynthia Garver of Oxford University Press and to Elaine Kehoe for ably shepherding the huge and complex manuscript of Reinventing Identities through the publication process. 1. Recent collections include those edited by Victoria Bergvall, Janet Bing, and Alice Freed (1996); Jennifer Coates (1997); Kira Hall and Mary Bucholtz (1995); Sally Johnson and Ulrike Meinhof (1996); Anna Livia and Kira Hall (1997); Sara Mills (1995); Sue Wilkinson and Celia Kitzinger (1995); and Ruth Wodak (1997). These are in addition to numerous journal articles, monographs, and textbooks that have recently been published or are in process. 2. There are, however, important differences in the details of these theories. Butler's skepticism about the universal category of the subject and the power of individuals in the face of hegemonic ideologies leads her to formulate her position in less active language than does Anzaldua. (It has been a cause for dismay among many feminists that in Butler's writings language [which here "inserts itself in ... linguistic life"] is often more agentive than individual subjects ["the resulting effects of ... discourse" (1990:145)]. Yet her theory of drag also offers the possibility of speakers' active resistance to normative cultural discourse) Furthermore, where Butler maintains that language constructs the self (via cultural discourse), Anzaldua implies that identities may exist prior to the language with which they are projected and displayed. 3. For details, see for example Marcyliena Morgan's (this volume) discussion of Claudia Mitchell-Kernan (1971); Bucholtz and Hall's (1995) rereading of Robin Lakoff (1975); Sara Trechter's (chapter 5, this volume) analysis of the work of Mary Haas (1944); and Caitlin Hines's (chapter 7) and Anna Livia's (chapter 17, both this volume) separate treatments of several "Whorfian" language and gender researchers. See also Patricia Nichols's pioneering sociolinguistic work on black women's language (e.g., Nichols 1978, 1983), which the field's patriarch, William Labov, did not cite until 1990, after Penelope Eckert (1989) discussed it

22

REINVENTING IDENTITIES

in an influential critique of his view of women's language. (My thanks to Keith Walters for this example.) 4. Not only has Eckert and McConnell-Ginct' s essay inspired research within the model, but it also has provided a theoretical framework through which to understand work already being done. Thus their theory, like most others, is part invention and part discovery. Their work is a valuable example of how language and gender scholars can produce, as well as use, feminist theory in their work. It is also important to note that Eckert and McConnell-Ginet's concept of practice differs from Butler's. For Butler, practices are culturally imposed ideological structures that assign subjecthood; for Eckert and McConnell-Ginet, they are the activities taken up by participants in local communities, possibly in complicity with and possibly in resistance to cultural ideologies of identity. REFERENCES

Althusser, Louis (1971). Ideology and ideological state apparatuses. In Althusser, Lenin and philosophy and other essays. Trans. Ben Brewster. New York: Monthly Review Press, 127-186. Anzaldiia, Gloria (1987). Borderlands /'La frontera: The new mestizo. San Francisco: Spinsters/ Aunt Lute. (ed.) (1990). Making face, making soul /Hacienda caras. San Francisco: Aunt Lute. Austin, .1. L. (1962). How to do things with words. Cambridge, MA: Harvard University Press. Bauman, Richard (1977). Verbal art as performance. Prospect Heights, JL: Wavcland Press. Bergvall, Victoria L., Janet M. Bing, & Alice F. Freed (eds.) (1996). Rethinking language and gender research: Theory and practice. London: Longman. Bodine, Ann (1975). Androcentrism in prescriptive grammar: Singular "they," sex-indefinite "he," and "he or she." Language In Society 4:129—146. Bourdieu, Pierre (1991). Language and symbolic power. Ed. John B. Thompson. Trans. Gino Raymon & Matthew Adatnson. Cambridge, MA: Harvard University Press. Briggs, Charles L. (1988). Competence in performance: The creativity of tradition in Mexicano verbal art. Philadelphia: University of Pennsylvania Press. Bucholt/., Mary, & Kira Hall (1995). Introduction: Twenty years after Language and Woman's Place. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 1-22. Butler, Judith (1990). Gender trouble: Feminism and the subversion of identity. New York: Routledge. Coates, Jennifer (ed.) (1997). Language and gender: A reader. Oxford: Blackwell. Dubois, Betty Lou, & Isabel Crouch (1987). Linguistic disruption: He/she, s/he, he or she, he-she. In Joyce Penfield (ed.), Women and language in transition. Albany: SUNY Press, 28-36. Eckert, Penelope (1989). The whole woman: Sex and gender differences in variation. Language Variation and Change 1:245-267. Eckert, Penelope, & Sally McConnell-Ginet (1992). Think practically and look locally: Language and gender as community-based practice. Annual Review of Anthropology 21:461—490. Fishraan, Pamela (1983). Interaction: The work women do. In Barrie Thome, Cheris Kramarae, & Nancy Henley (eds.), Language, gender, and society. Cambridge, MA: Newbury House, 103-118. Foucault, Michel (1980). Language, counter-memory, practice. Kd. Donald F. Bouchard, trans. Donald F. Bouchard & Sherry Simon. Ithaca, NY: Cornell University Press. Gaudio, Rudolf P. (1997). Not talking straight in Hausa. In AnnalJvia & Kira Hall (eds.), Queerly phrased: Language, gender, and sexuality. New York: Oxford University Press, 416-429. Goffman, Erving (1959). The presentation of self in everyday life. Garden City, NY: Doubleday.

Bad Examples

23

Grice, H. Paul (1975). Logic and conversation. In Peter Cole & Jerry L. Morgan (eds.), Syntax and semantics. Vol. 3: Speech acts. New York: Academic Press, 41-58. Haas, Mary (1944). Men's and women's speech in Koasati. Language 20:142-149. Hall, Kira, & Veronica O'Donovan (1996). .Shifting gender positions among Hindi-speaking hijras. In Victoria L. Bergvall, Janet M. Bing, & Alice F. Freed (eds.), Rethinking language and gender research: Theory and practice. London: Longman, 228—266. Harding, Sandra G. (1991). Whose science? Whose knowledge?: Thinking from women's lives. Ithaca, NY: Cornell University Press. hooks, bell (1981). Ain't la woman? Black women and feminism. Boston: South End Press. Hull, Gloria T., Patricia Bell Scott, & Barbara Smith (eds.) (1982). Hut some of us are brave. New York: Feminist Press. Hymes, Dell (1981). Breakthrough into performance. In Hymes, "In vain I tried to tell you": Essays in Native American ethnopoetics. Philadelphia: University of Pennsylvania Press, 79-141. (Original work published 1975) Irigaray, Luce (1985a). Speculum of the other woman. Trans. Gillian C. Gill. Ithaca, NY: Cornell University Press. (1985b). This sex which is not one. Trans. Catherine Porter with Carolyn Burke. Ithaca, NY: Cornell University Press. Johnson, Sally, & Ulrike Meinhof (eds.) (1996). Language and masculinity. London: Blackwell. Johnstone, Barbara (1996). The linguistic individual: Self-expression in language, and linguistics. New York: Oxford University Press. Johnstone, Barbara, & Judith Mattson Bean (1997). Self-expression and linguistic variation. Language in Society 26:221—246. Kulick, Don (1996). Gender in the speech of Brazilian transvestite prostitutes. Paper presented at the annual meeting of the American Association for Applied Linguistics, Chicago. Labov, William (1972a). Language in the inner city: Studies in the Black English Vernacular. Philadelphia: University of Pennsylvania Press. (1972b). Sociolinguistic patterns. Philadelphia: University of Pennsylvania Press. (1990). The intersection of sex and social class in the course of linguistic change. Language Variation and Change 2:205-254. Lakoff, Robin (1975). Language and woman's place. New York: Harper & Row. Lave, Jean, & Etienne Wenger (1991). Situated learning: Legitimate peripheral participation. Cambridge: Cambridge University Press. Livia, Anna, & Kira Hall (eds.) (1997). Queerly phrased: Language, gender, and sexuality. New York: Oxford University Press. Longino, Helen E. (1990). Science as social knowledge: Values and objectivity in scientific inquiry. Princeton: Princeton University Press. Mills, Sara(ed.) (1995). Language and gender: Interdisciplinary perspectives. London: Longman. Mitchell-Kernan, Claudia (1971). Language behavior in a black urban community. Berkeley, CA: Language Behavior Research Laboratory. Nichols, Patricia C. (1978). Black women in the rural South: Conservative and innovative. International Journal of the Sociology of Language 17:45—54. (1983). Linguistic options and choices for Black women in the rural South. In Barrie Thorne, Cheris Kramarae, & Nancy Henley (eds.), Language, gender, and society. Cambridge, MA: Newbury House, 54-68. Parker, Andrew, & Eve Kosofsky Sedgwick (eds.) (1995). Performativity and performance. New York: Routledge. Penelope Stanley, Julia (1977). Paradigmatic woman: The prostitute. In David L. Shores & Carol P. Hines (eds.), Papers in language variation: The SAMLA-ADS collection. University: University of Alabama Press, 303-321.

24

REINVENTING IDENTITIES

Schulz, Muriel (1975). The semantic derogation of women. In Barrie Thorne & Nancy Henley (eds.), language and sex: Difference and dominance. Rowley, MA: Newbury House, 64—75. Taylor, Anita (ed.) (1997). Women and Language, 20(1). Rethinking gender. [Special issue], Thorne, Barrie, Cheris Kramarae, & Nancy Henley (1983). Language, gender and society: Opening a second decade of research. In Thorne, Kramarae, & Henley (eds.), Language, gender, and society. Cambridge, MA: Newbury House, 7—24. Weir, Allison (1996). Sacrificial logics: Feminist theory and the critique of identity. New York: Routledge. West, Candace, & Don H. Zimmerman (1983). Small insults: A study of interruptions in cross-sex conversations between unacquainted persons. In Barrie Thorne, Cheris Kramarae, & Nancy Henley (eds.), Language, gender, and society. Cambridge, MA: Newbury House, 103-118. Wilkinson, Sue, & Celia Kitzinger (eds.) (1995). Feminism and discourse: Psychological perspectives. London: Sage. Wodak, Ruth (ed.) (1997). Gender and discourse. New York: Sage.

Part I

IDENTITY AS INVENTION

This page intentionally left blank

1

MARCYLIENA MORGAN

No Woman No Cry Claiming African American Women's Place My work requires me to think about how free I can be as an African-American woman writer in my genderized, sexualized, wholly racialized world. To think about (and wrestle with) the full implications of my situation leads me to consider what happens when other writers work in a highly and historically racialized society. For them, as for me, imagining is not merely looking or looking at; nor is it taking oneself intact into the other. It is, for the purposes of the work, becoming. My project rises from delight, not disappointment. It rises from what I know about the ways writers transform aspects of their social grounding into aspects of language, and the ways they tell other stories, fight secret wars, limn out all sorts of debates blanketed in their text. And rises from my certainty that writers always know, at some level, that they do this. (Morrison 1992:4)

oni Morrison wrote the above passage to explain her need to enter the volatile terrain of criticism concerning African presence in U.S. literature. My analogous purpose, as an African American female linguist, concerns the racialized, sexualized, and genderized context of linguistic research. Any exploration of how language constitutes, instantiates, and constructs the social world is fraught with assumptions about race, class, gender, and sexuality that are conventionally accepted under the rubric of "knowledge," "findings," and "objectivity." Under the guise of science we, especially women, people of color, and women of color, are silenced or adroitly handled—as in manhandled—especially in work on the language of African Americans and the marginalization of women within such work. This discussion, like Morrison's, rises from delight rather than disappointment. The issue is not to bemoan the past but to participate in and agitate for a present and future that are socially inclusive and intellectually productive. Because language is a social act, research on language constitutes social and cultural production that is influenced by issues of race, sexuality, class, and power. Thus the marginalization of women in African American language research is not about gender exclusively. One way to begin is to examine the representation of African American women's language use while pondering the challenges Black women face daily. Perhaps the most significant challenge, and one that biases all scholarly research, concerns how 27

28

IDENTITY AS INVENTION

African American women are viewed in relation to others, especially Black men and White women—how their identities are assigned as part of a system of dichotomies rather than discovered as something much more complex. The exclusion and marginalization of Black women is not limited to scholarly research. In the legal system too, according to Kimberlc Crenshaw (1992), race, class, and gender consistently intertwine in African American women's lives. She argues that Black women experience intersectionality in most legal decisions concerning race or gender. For example, Black women who sue employers for sexual harassment can seldom include White female workers in their suits. Similarly, they are challenged when they attempt to include Black men in racial discrimination cases involving Black women. Consequently, African American women's issues are hypermarginalized and are considered typical neither of all women's issues (because the women who face them are Black) nor of Black issues (because the Blacks who face them are women). It is not surprising, then, that all linguists—whether they include, marginalize, or fetishize Black women—always, at some level, take a position on this situation. The position taken here is that African American women participate in the development of language norms, the introduction of innovations, and the use of all varieties of African American English (AAE).1 Moreover, their contribution is most apparent in cultural settings in which they are social actors—in places, that is, where identity is central. The issue of African American women's identity, and women's identity in general, is shrouded in postmodernist discussions that are seldom based on the choices and challenges of everyday life. In fact, ethnographic and linguistic descriptions are often summarily dismissed as essentialist if they do not apply to all cultures and to all issues of globalization. Yet within postmodernist theorizing women continue to be essentialized. Stuart Hall both critiques and continues this practice when he applauds feminists' rejection of the Cartesian and the sociological subject: "Feminism challenged the notion that men and women were part of the same identity—'Mankind'—replacing it with the question of sexual difference" (1995:611). Although this description may suit Western academic feminism, it does not begin to address the complex ways women throughout the world (and in the West itself) experience and theorize their identity as women. Moreover, African American women's identity exists in relation to White women's identity, perhaps more so than to (Black) male identity. Black women are presented as problematic with respect both to feminist issues and to patriarchal ideals of White womanhood. The majority of Black women, after all, are workers who also have authority at home, a reality still unrecognized in White feminist political agendas. Thus Judith Butler's admonition, "Reading identities as they're situated and formed in relation to one another means moving beyond the heuristic requirement of identity itself" (1995:446), is appropriate, especially given American feminists' defining moments of gender and race in the Black community during the mid-1990s: the Clarence Thomas Senate hearing and O. J. Simpson's murder trial. Both cases heightened awareness that race and gender are not interchangeable concepts for Black women but rather fused. These events are significant not because they illustrate a racial split within feminism but because they forced all U.S. women to consider the difficulty of juggling issues of racial and sexual discrimi-

No Woman No Cry

29

nation in reality rather than in theory. Notwithstanding the resulting acrimony, White feminists and feminists of color have a great deal in common. As bell hooks (1990, 1992), Kamala Visweswaran (1994), Dorinne Kondo (1997), and others challenge, feminist theory must be situated at home, a place we have been before but never really experienced. To explore this place is crucial, for the intersections of these factors greatly affect linguistic analysis in general and descriptions of language use among African Americans, women, and African American women in particular. The following discussion is a description and reanalysis of African American women's language and cultural practice across generations. The interviews are based on long-term ethnographic fieldwork with Black women in the United States and other parts of the English-speaking African diaspora. With few exceptions, women who have been socialized within African American culture are familiar with three central interactional events: children's he-said-she-said interactions; teenagers' and young adults' instigating; and adult women's conversational signifying. These occur among each generation as part of a process of social face and identity construction. Previous study of these practices as unrelated to other cultural activities missed girls' significance in the description of the African American community (Goodwin 1990 is an exception) and the importance of social face among women.

The vernacular as male and poor The concept of the vernacular is central to sociolinguistic study of African American language. Sociolinguistics uses quantitative methods to analyze linguistic variables: that is, structural items that occur frequently in conversation and whose frequency of occurrence is highly stratified according to age, class, and other factors (Labov 1972:8). Thus a study of the vernacular, the ordinary language of a people, implies analytic focus on everyday activities and social actors. Yet many descriptions of African American speech have been based on data from adolescent boys in research interviews (cf. Morgan 1994a; Mufwene 1992). Until recently, the description of vernacular or AAE speakers effectively marginalized African Americans who did not fit the stated criteria, such as "black youth from 8 to 19 years old who participate in the street culture of inner cities" (Labov 1972:xiii). In particular, linguists too often view language varieties of adolescent male gangs as authentic or core AAE, as when William Labov contrasts his core (gang) group with the "lames," young Black men who accept parental influence, attend school regularly, pursue the advantages of the dominant culture, are inactive because of poor health, or are mentally or morally defective: that is, punks (1972:259). This description of vernacular culture constructed "authentic" African American membership, identity, and language as male, adolescent, insular, and trifling. Everyone else is a "lame." Since lames do not participate in core culture, do not use AAE features in the same ways, and speak another version of AE (Spears 1988), researchers do not view them as culturally African American. Moreover, where the Black community is seen as vernacular-speaking by definition, most studies consider Whites to be speakers of Standard English (cf. Morgan 1994b, c). Labov (1998) addresses these issues

30

IDENTITY AS INVENTION

by distinguishing between American varieties under the headings of General English (GE), Other American Dialects (OAD), African American (AA), and African American Vernacular English (AAVE). The confusion about African American vernacular language, especially its association with males, can also be seen in cultural criticism. Mary Helen Washington (1987) argues that Janie, in Zora Neale Hurston's novel Their Eyes Were Watching God (1937), uses folk language (dialect) to symbolize the limit of her power but that the Reverend John Pearson's use of folk language in Hurston's Jonah's Gourd Vine ([1934] 1990) leads to liberation. Washington condemns Hurston's differential treatment of her characters, but she equates dialect with maleness, concluding that it is limiting for the woman and liberating for the man. However, the vernacular is the likely variety for both characters. When scholarly notions of the vernacular define everyday speech as male speech, women are not integral to generalized definitions and descriptions of speakers. It is not always clear whether women are even included in sociolinguistic data analysis. Still more alarming, the definition of the vernacular reifies stereotypes of African American culture: young men with nothing to do, doing nothing, talking trash, and going nowhere (Duneier 1992). Yet if we reclaim the general definition of the vernacular as core culture, as John Gwaltney (1981) does, and include women, men, and children along with adolescent boys, the vernacular becomes representative of "regular" speech, and women's voices can be heard. That is, core culture is not a question of an authentic subject but of one who lives within the social norms and cultural practices of a multiclass, intergenerational, and gendered African American culture. Language and gender Though often excluded or marginalized in language studies (cf. Eckert & McConnellGinet 1992; Henley 1995), African American women are occasionally mentioned. For example, Roger Abrahams (1970) remarks, with no explanation, that African American women refused to participate in his Philadelphia folklore project. He later describes them as not participating in verbal play and as "restrained in their talk, less loud, less public, and much less abandoned" compared with men (1974:242). Finally, in his examination of the representation of women's speech styles in literature, he suggests that women may have the same expressive acuity as men (Abrahams 1975). Other reports of African American women's language use, while scarce, comment on their linguistic conservatism; their role as the "real" target and often the audience, observer, and supporter of male signifying games; and their willing collaboration in street encounters (e.g., Abrahams 1962; Kochman 1972; Labov 1972; Wolfram 1969). Claudia Mitchell-Kernan (1971) produced one of the few early works that did not describe urban African American women in relation to men and as aggressive, domineering, and emasculating. Her rich ethnography demonstrated that women participate in conversational signifying and employ other linguistic practices similar to those of men. Yet Mitchell-Kernan's work has been criticized because of her gender—she is reprovingly described as "a young attractive Black

No Woman No Cry

31

woman"—and her "middle-class" status (Kochman 1973:969,970). In a vindication of Mitchell-Kernan's work, Henry Louis Gates Jr. (1988) relied heavily, point for point, on her description of signifying as a foundation for his theory of African American discourse. It is also important to remember that before the late 1980s and the implosion of the margins into the center of academic inquiry (Kondo 1997) with critical work by feminist and minority scholars, numerous works by women, natives, and Others were summarily dismissed as too personal and subjective (Visweswaran 1994). Fortunately, the scholarship on women's language use in their communities is growing. Current research critiques the prevailing literature on African American women's and girls' speech (e.g., Ball 1992; Etter-Lewis 1991, 1993; Etter-Lewis & Foster 1996; Foster 1995; Goodwin 1980, 1982, 1985, 1988, 1990; Morgan 1989, 1991, 1993). Similarly, much of John Rickford's work, as well as his collaborations with colleagues (Rickford 1986; Rickford, Ball, Blake, Jackson, & Martin 1991; Rickford & McNair-Knox 1993) on variation and style-shifting in AAE, is based on long-term interviews with a young female community participant. Yet women and girls are still often compared with their male counterparts, which clouds both commonalities and gender differences. A vivid example can be found in the confusion over the significance of a cool social face.

Face to face: Staying cool One important aspect of African American culture that equally affects women and men is the maintenance of the speaker's social face, an impression formed of a person based on her or his self-presentation (Goffman 1967). Perhaps the most widespread African American cultural concept that both critiques and symbolizes social face is the notion of being cool—current and trendsetting, calm, detached, yet in control (cf. Major 1994; Smitherman 1994). Gwaltney pays tribute to this cultural value in describing one of his community contributors, Nancy White: "She is the exemplar par excellence of the highest status that core black culture can accord— that of the cool, dealing individual" (1981:143). A cool face is the ability to enact subtle symbolic cultural practices with eloquence, skill, wit, patience, and precise timing. Although some commentators have tied cool social face to racism or male coping skills (Abrahams 1962; George 1992; Grier & Cobbs 1968; Horton 1972; Kunjufu 1986; Majors & Billson 1992), coolness is mainly a cultural practice (cf. Gwaltney 1981; Smitherman 1977) used by both genders and having counterparts in all parts of Africa and the African diaspora (Alleyne 1980, 1989; Yankah 1991 a, b).2 It contrasts with fools or acting a fool—an insult that both denigrates and dismisses a person as a cultural member, as Mabel Lincoln, another woman in Gwaltney's study, describes: "To black people like me, a fool is funny—you know, people who love to break bad [i.e., to suddenly behave as though one is knowledgeable or an authority], people you can't tell anything to, folks that will take a shotgun to a roach. . . . But most of us try to be cool. That is what we respect the most in ourselves and look for in other people. That means being a person of sober, quiet judgment" (Gwaltney 1981:68-69).

32

IDENTITY AS INVENTION

Being cool is especially important in interaction, because discourse requires that all participants (including hearers) constantly assess and address potential meanings within and across contexts. A cool social stance is multiply constructed through social context and contrastive use of AAE and AE. As a symbolic "good" with exchange value, coolness can be used to accrue linguistic and pragmatic capital. Individuals can "lose their cool" or social face in interactions in which they are culturally challenged (for example, by not knowing current lexical terms or meanings) or in which the dominant culture (the police, the legal system, the school, and so on) claims not to understand a particular form of interaction, such as indirection. A third interviewee of Gwaltney's, Mrs. Briar, describes herself as a little girl learning how to be cool and the penalty to one's social face resulting from acting a fool. I was five when I learned not to lose my cool when the trucks backfired. I remember the day it happened. I had asked my father for something and he had said no. But when the trucks came by and backfired, 1 just sat there like nothing had happened. He said, "Girl, let me shake your hand!" and he gave me money and I felt just as tall as he was. My brother Harry, who is three years older than I am, without even looking at me said, "No cool." Then everybody teased him for running off at the mouth without knowing what he was talking about and he felt bad, I think. . . . We don't like to show out, but if you guess wrong, you might be, well, you might be out there all by your lonesome. (Gwaltney 1981:192)

This passage shows the role of explicit socialization practices in teaching children the importance of coolness. African American children also learn how to be cool through language play and play language, especially the ritual games of signifying and instigating.

Children's language play Although African American children play numerous language games (Goodwin 1990), the one that has received the most scholarly attention is signifying, a verbal game of indirection also known regionally as sounding, the dozens, joning, snapping, busting, capping, bagging, and ranking (Abrahams 1962; Garner 1983; Gates 1988; Kochman 1972; Labov 1972; Mitchell-Kernan 1972,1973; Percelay, Monteria, & Dweck 1994; Smitherman 1977). Mitchell-Kernan describes signifying as "the recognition and attribution of some implicit content or function which is obscured by the surface content or function" (1972:317-318). It is a form of play for adolescent boys in particular, and it can serve indirect functions in adult interaction, as in conversational signifying. Many have suggested that signifying started as an outlet for racial oppression (e.g., Abrahams 1962; Dollard [1939] 1973; Kochman 1972; Percelay et al. 1994) because one must maintain one's cool in order to play. However, verbal coolness is a culturally constructed form of social face, and its obvious function as an outlet is probably an added bonus for youth who must learn both the cultural rules of the Black community and the political reality of being Black in America.

No Woman No Cry

33

The notion of play in signifying differentiates the real from the serious (Abrahams 1970, 1976; Goffman 1974; Kochman 1983, 1986) by placing socially or culturally significant topics (e.g., personal details, including one's mother, relatives, sexuality, physical appearance, class, and economic status, as well as cultural common ground such as political figures) in implausible contexts. Plausibility is partially determined by cultural values. Thus a signifying episode that places one's mother in a possible context such as being unemployed or pregnant could be considered an insult and lead to physical confrontation. Conversely, the notion of a police officer who "serves and protects" the Black community would be considered implausible and hence could be a resource for signification. Once the implausible or unreal state is established, cultural signs interact with context through irony, sarcasm, wit, and humor in order to play with the serious signifier. For example, one commonly heard signifying turn is "You're so ugly, you went into a haunted house and came out with a job application." If the sign fits the context (that is, if you are ugly), the interaction is considered an insult rather than play. As verbal play, signifying or snapping is mainly performed by adolescent boys, although it also occurs among adult women and men involved in competitive activities such as sports or stock trading. Signifying may occur as routine play among boys, and losing one's cool within the parameters of the game is an indication that the player has lost. Winners and losers are determined by the onlookers. If a player loses, he may redeem himself during the next day's play. Thus his social face as a cool-headed individual can be retrieved. In contrast, girls' language activities have more inherent risks because they are not bounded as play. For example, a girl who talks about another girl behind her back risks being labeled an instigator, and unlike her male peer who plays signifying games, she cannot redeem herself during the next day's play. She must undergo an elaborate waiting game and reconciliation session before reestablishing herself among her peers. Girls' verbal activities focus on the content of previous and future interactions. Goodwin's (1980, 1990, 1992) analysis of he-said-she-said disputes among African American girls illustrates the elaborate lengths to which participants are willing to go in order to determine who said what behind someone's back. The role or motive of the instigator is not generally investigated by the offended party (Goodwin 1992), who works to maintain or reestablish her social face by telling narratives projecting her own future action in response to instigating stories.3 This response is expected; as Goodwin (1992:187) explains, "The goal of the instigator's storytelling is to elicit a statement from the offended party which leads to her confronting the offending party." Younger girls maintain social face by demonstrating that if someone is suspected of talking about another girl behind her back, the offended party will investigate the action rather than the intention of the instigator and then confront her, sometimes physically. As girls get older, however, they shift their focus to include the intentions of the instigator and other participants and whether these correspond to the offended party' s interpretations. Consequently, talk increases among all participants. Moreover, talking about someone behind her back takes on a new seriousness: The activity is not simply gossip but rumor. In the African American speech community, when a rumor achieves widespread audience discussion and assessment, it is often treated as truth,

34

IDENTITY AS INVENTION

even when it is not believed to be factual (cf. Turner 1993). And because a rumor also signals a loss of social face, its target must defend her honor. By the time girls have become teenagers, they have a significantly different focus on the instigator in he-said-she-said events. Young African American women treat talking behind another's back with the same seriousness as a capital offense. Before the alleged offending party is confronted, the accused party must prove that the intermediary who reported the offense is not simply an instigator. When the offending party is ultimately confronted, she may avoid physical confrontation if she admits to starting the rumor and apologizes or if she convinces the offended party that her intentions were misunderstood. Often the offending party admits to making the incriminating remark but does not apologize, or she refuses to admit that she said anything, although others report that she did. In both cases, some sort of directed confrontation is possible. Instigating events are therefore about participants and occurrences of talk, as well as about what was allegedly said by whom. For teenagers, the event is designed to expose and either acquit or convict the instigator and the offending party. Days or weeks may elapse as statements are denied or confirmed and analyzed by witnesses. The offended party's aim is to determine who started the rumor. In the process, friendships are tested, conversational roles are assessed, and all parties become invested in identifying the alleged perpetrator of the offending speech event. As with young children, the person reported to be the source of the statement is the last person contacted. The following story told by Zinzi, a 20-year-old college student, describes an instigating episode initiated by Sheila against Zinzi when both girls were in high school. Zinzi told this story in an undergraduate class after reading Goodwin's (1990) book detailing he-said-she-said interactions. She introduced the episode amid joking from classmates (who had similar stories) that the instigator had told the truth; Zinzi then adopted a defensive posture, her head and eyes slowly rolling, and told the story.4 (1)

1 Zinzi: 2 3 4 5 6 7 8 9 10 11 12 J3 14 15 16 Morgan:

. . . And then so she thought that she was close enough to Tyrone and so Tyrone wouldn't tell me. BUT? Tyrone being the BEST friend that he is, he's just like, "You know? Sheila is spreading ru?mors about you. I don't know if anybody else? told? you, but you know, she saying that you and Barry been DOing things and duh?duhduh.dahdah.dah?" And I was just like (2) "Oh? she di:d? huh?" And then so I decided (2) just instead of going up in her face—'cause I didn't like her anyway— instead of going up in her face, that I'd go and ask my OTHer friends and things like that. So I went and asked them, and they were like, "Yeah, yeah, ((high-pitched, soft voice)) she did tell me about that but I didn't believe her." And I'm like—Uh huh, yeah, right! That's how come you didn't TELL me, because you didn't BELIEVE her. Yeah (2) okay And so an?yway, when I went and confro:nted? her. And then I just got the satisfac'.'tion out of it (2) because all it took was like a little? confrontation and What did you say?

No Woman No Cry 17 Zinzi: 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Morgan: 37 Zin/J: 38 39

35

Well 1,1 asked her?—well not ACTually ASked her—but I accused? her? and I was like "Oh, so I heard that you been telling ru?mors about Barry and I." And then she? (2) didn't deny? it. And she was just like "It DID? happen." And I'm like "How do you know it happened then?" So, at first? we were talking? lo:w? and then got kind? of lou::d and the:n? since this was like in front of the church? house. And then it was like, okay (2) let's just take this ELSEwhere. And you KNOW how when HIGH school kids get—just like (2) when you TAKE stuff elsewhere and then EVERYBODY! FOLLOWS. And then it's like (2) ALRIGHT (2) now I'm going to have to fight her 'cause EVERYbody else is over here too. And then so she was still talking her little SMACK LIP? (2) and things like that. And you know (2) everybody was like "Yes you DI::D say that (2) and I HEARD IT" and she was like "Yeah 1 DI::D say it because it IS TRUE?" And I'm just like "You DON'T know NOTHING about NOTHING and dah?dahdahdahdah? And then so—((suck teeth)) that was it (2) when she just got up in my face. And I could just (2) SMELL her breath? and FEEL her spit? and it was just like ((claps)) tat! And it was on ((laughs)). Wait a minute. No! You fought? Of course, ((laughs)) Like, what did you WANT me to DO?: "Well that's okay you can? go ahead and tell rumors about me? Go right ahead" ((in a hypercorrect, high-pitched voice)). No!

In contrast to boys' signifying episodes, girls' instigating often leads to physical confrontations, which are not viewed as a loss of face or coolness but as a logical last resort. Zinzi confirms this in lines 7 and 8, where she reports deciding to ask her friends what Sheila said "instead of going up in her face."5 Zinzi's story is different from those reported by Goodwin in one major respect: The basic three stages Goodwin (1990, 1992) describes—(1) offense, (2) instigating, (3) confrontation—have been expanded to include interrogation of "so-called" friends and punishment of the offender.6 The social order is clearly in jeopardy as Zinzi canvasses her friends for their role in Sheila's conversations in lines 9 through 11. She determines who her "real" friends are and whether Tyrone's report is true by interrogating friends and bystanders. Zinzi then finds out exactly how her friends responded to Sheila. During this time, all parties focus on past, present, and possible conversations with and about Sheila. Once she has established the truth, Zinzi searches for, finds, and confronts the instigator. Sheila accepts responsibility for what she says (lines 19-20 and 29-30), Zinzi denies that it is true, and eventually she strikes Sheila. She explains her physical attack on Sheila by defending her right to protect herself against unfounded rumors. Although all of these children's language activities are fraught with confrontations and accusations, they also illustrate the construction of social role and relationship through indirectness, cultural symbols, and audience coconstruction and collaboration. Participants desperately maintain a social face and the respect that it

36

IDENTITY AS INVENTION

entails. This is true for both female and male children and adolescents, although girls, through their elaborate procedures, are much more active in maintaining social and cultural rules. Children arc invested in and aware of the multilayered nature of such activities, especially how speaker intent and meaning arc constructed and validated. Signifying and instigating also occur in adult conversations, though in slightly different forms. Adults maintain and often expand the level of complexity common in childhood but have a very different attitude about how to play with available language styles and varieties.

Adult interaction: Reading dialect A competent adult speaker already knows what children eventually learn through signifying and instigating routines: that audience and hearers are equal partners in interpreting talk. The Black expression "Know when you're playing" (also Gwaltney 1981 :x) highlights the importance of audiences in constructing meaning in conversation. Play refers to the speaker's intentionality, as well as to whether she or he tells the truth and understands the consequences of what is said. This is an important characteristic of adult interaction, because in African American culture intentionality is considered not a psychological state but a collaborative social construct (cf. Duranti 1993). The coconstruction of meaning and intention is tied to both language and culture in adult interaction. Unlike children, adults exploit the linguistic and pragmatic resources of both AAE and AE to constitute, construct, and sometimes expose social relations and cultural knowledge and norms. An example of this process involves a practice I term reading dialect (Morgan 1994a), which should be understood in terms of the more general African American interpretive practice of reading, a form of directed discourse in which a speaker unambiguously denigrates another to her or his face (Goffman 1967).7 Although one may self-report having read another person without witnesses (for example, in telling a story, the narrator may simply report, "I READ her!"), reading is legitimately accomplished only in the presence of other witnesses who corroborate that it occurred.8 Reading is directed, often accusatory, speech. When a target is read, her or his social face is attacked for inappropriate or offensive statements or for what the reader perceives as the speaker's false representation of her or his beliefs, personal values, and so forth. One may be read for acting out class privileges, failing to greet friends, or misrepresenting one's beliefs (Morgan 1996). The point is not the reader's correctness but the willingness to jeopardize her or his own social face by disclosing, regardless of setting or context, the target's perceived attempt to camouflage personal beliefs, attitudes, and so on.9 Reading dialect, a subgenre of reading, occurs when members of the African American community contrast or otherwise highlight obvious features of AAE and AE to make a point. The point is not necessarily negative, but the grammatical contrast indicates a challenge to someone's social face. These lexical and grammatical structures are well known in the community and are often the focus of verbal play, humor, and irony. For example, to stress a point a member might say, "It's not simply that I am cool. I be cool. In fact, I BEEN cool (a very long time)." In the African

No Woman No Cry

37

American community, not only the two dialects of AAE and AE but also varieties within those dialects are regularly read by interlocutors. Reading dialect often highlights the social-face contrast of being cool versus acting a fool in that the reader challenges the target's discourse. This challenge is also in defense of the reader, who does not want to be taken for a fool and thereby incur damage to her or his social face. Discussions between the reader and the target often involve clarification, agreement, and disagreement, ending with social face intact and opposing statements from the target and the reader: The target states that the reader's interpretation is wrong and the reader states that it is correct. African American women in (inter)action As African American girls grow into women, their everyday conversations often involve the expression and defense of social face. Active participation in discourse is often based on extent of personal involvement in the events being discussed. If all major participants are not present, especially when another's speech may be reported, the event is only minimally discussed, even when participants are close friends or relatives. The origins of this practice lie in the importance adult women attach to the audience's right to determine intentionality, even when information is delivered behind the speaker's back. Women learned the crucial role of the audience when, as teenagers, they went to elaborate lengths to determine what someone said and how others responded to rumors. These interrogations are filled with accusations, and people must prove their friendship. Within women's interactions, however, the main discourse focus is not whether someone instigates or backstabs but rather whether the intentionality assessments made by the audience are reasonable considering the context and whether the original speaker had the opportunity to address these assessments. Thus, instead of focusing on who said something negative about another, as preadolescent girls do, or on who intended to start a confrontation, as teenagers do, women focus on a speaker's right to be present to represent her own experience. This right is fiercely protected, for it provides the conditions for the more fundamental right that women and men should be allowed to interpret their own experiences. Yet adult women's social face is even more delicately constructed because it is continually challenged and tested by the audience. Women operate with two dialogic styles, "behind your back" and "to your face," as represented by the statement "1 wouldn't say anything behind your back that I wouldn't say to your face," which is often used to challenge someone's social face and can halt he-said-she-said or instigating attempts. A woman who makes such a statement is viewed as standing up for what she believes in and says, irrespective of the costs. Not surprisingly, then, interactions about people who are not present are considered tactless and divisive. Therefore, talking about someone behind her back does not mean that the speaker says something derogatory but that the teller's intentionality or actual words do not have the benefit of coauthorship, and as a result the interaction may be misunderstood. The "behind your back / in your face" dichotomy stipulates that intentionality is socially constructed and anyone who subverts this construction intends to deprive others of their discourse rights.

38

IDENTITY AS INVENTION

The following conversational segment is illustrative. Participants include three related women who grew up together: Ruby (a jazz musician, age 78), who does not speak in this segment; Baby Ruby (a retired prison guard, age 63); and Judy (a retired data-entry worker, age 63). Ruby and Judy are sisters and Baby Ruby is their niece. Baby Ruby is not happy that she is still called by her childhood name, and she is not happy that Ruby and Judy are her aunts, a fact she laments to anyone who will listen. Also present are Judy's six daughters (including me). Other than my attempts at questioning, no daughters participate in the conversation, because for African American women and girls, mere presence during a conversation does not authorize participation. Girls have two expressions for uninvited conversational participants: "This is an A and B conversation so C your way out" and "You're all in the KoolAid and don't even know the flavor." (2)

"Auntism" 1 Morgan: NUMBER ONE uh—the First! question is: 2 (.) 3 now: in terms of growing up: right, you two were born 4 (.) 5 same year? right 6 (.) 7 Baby Ruby: =Six months apart and I'm in [I'm 8 Judy: [And she NEVER let me forget it.= 9 Morgan: =((laughs)) 10 (.) 11 Baby Ruby: Right 12 (.) 13 Baby Ruby: [But I 14 Judy: [She's SIX months older than I am 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

Baby Ruby: Judy: Baby Ruby: Judy: Morgan: Judy: Morgan:

Baby Ruby: Morgan:

(•) But that's the aunt. (.) And I AM her aunt. (.) And I:: don't like it. (.) And I:: don't care= =((laughs)) (.) I am STILL the aunt (.) NOW: you have to understand we never knew:: (.) that—you were her—she's your aunt (.) [YOU—you's [WE WERE AL:WAYS:! confused?

No Woman No Cry 33 34 35 36 Baby Ruby: 37 38 Baby Ruby: 39 40 41 42 Judy: 43 44 Baby Ruby: 45 Morgan: 46 47 Judy: 48 49 Judy: 50 51 Judy:

52 53 54 55 56 57 58 59 60 61 62 63 64 65

Baby Ruby: Morgan:

Baby Ruby: Judy: Judy: Baby Ruby: Baby Ruby:

39 (.) Yeah we—we were like what's the reLA:tionship (.) ((gazes at Morgan)) You're KIDDIN? (.) That's my DAD'S si:ster ((nods head toward Judy)) (.) Ain't THAT disGUSTin? (.) Your bad what? (.) [My DA::D'S sister [My DA::D'S sister (.) Right. (.) I AM her father's sister ((winks at granddaughter/camera)) (.) My dad- father- And uh:: she- I don't know why: you all didn't know it because she AL:ways sa::id: that I'm [six months ol:der than you [I SURE DID! (.) Well YEAH-But you-Yeah-I'm six months older than you:: than you doesn't mean:: [that [AH—DO—AND YOUR AUNTISM DOESN'T GO ANYWHERE [And she'd always call me (?) (.) She [A:Lways said it [CAUSE I'M THE OLDEST (.) So your auntism: is: like nothing?

In "Auntism," Judy and Baby Ruby offer competing perspectives on their relationship. In the process, they talk "in each other's faces" about each other, using hearers (Judy's daughters) to mimic talking about someone behind her back. Lines 1 through 14 initiate an interactional sequence in which Baby Ruby and Judy respond to my question about their being the same age, which for them is also a kinship question. This interaction quickly becomes a competition over who will tell the story: Judy overlaps Baby Ruby (line 8) and completes Baby Ruby's point while overlapping with Baby Ruby again in line 14. Beginning in line 16, Baby Ruby and Judy argue about their kinship, addressing each other and their daughters and granddaughters, who function as mock receivers and overhearers. Judy and Baby

40

IDENTITY AS INVENTION

Ruby do not use direct eye contact with each other, although they do manage a few sideways glances. Baby Ruby and Judy signify on each other by reading dialect and using mock receivers. In particular, Baby Ruby signifies through reading dialect in line 16 when she invokes the unambiguous AAE usage of the demonstrative pronoun that to refer to an animate entity, namely Judy, in order to convey a negative reading. In AAE that is frequently used to emphasi/e that a person is the target of signifying. In these contexts that if, marked negatively because many members of the African American community, especially older members, interpret use of an inanimate term in reference to a Black person as insulting, regardless of the speaker's race. That bears additional significance because many older African Americans were raised in the South, where white supremacists referred to Black adults as children or objects. Baby Ruby directs her statement about Judy, Bui that's the aunt, to me (the mock receiver). Judy signifies back by also directing her comment to me and by reading dialect with the first-person Standard English noncontracted copula AM spoken loudly in line 18. AM is spoken as part of loud-talking, because it is noticeably louder than preceding and following utterances (cf. Mitchell-Kernan 1972). It thus marks the claim made in Standard English as authoritative: Judy is the aunt. This turn also begins a series of contrasting parallel statements that are conjoined by and (lines 18-22), which are part of signifying because their rhythmic similarity highlights their contrasting lexical and grammatical relations. Line 16 begins the assessment dispute over the nature of the technical definition of aunt and the term's associated social norms. In line 20 Baby Ruby offers her subjective negative assessment of Judy's being her aunt. Judy responds with a parallel structure in line 22, a negative comment regarding Baby Ruby's statement, and in line 25 she mirrors line 18, with the adverb still highlighting the fact that although Baby Ruby doesn't like it, Judy will always be the aunt. However, the dispute over who has the right to define their relationship has not ended. Although Judy's topic change interrupts the signifying episode (line 51), Baby Ruby has not finished asserting her right to define the relationship. In line 58, she further diminishes Judy's status by recasting Judy's repeated assertion—I AM her aunt; 1 am STILL the aunt—as "YOUR AUNTISM." She changes the quality of the noun aunt by adding the suffix -ism, which denotes the attitude, role, and responsibilities of being an aunt (cf. Quirk et al. 1972). Thus Baby Ruby replaces Judy's formal definition of their relationship with her notion that Judy never had the duties, responsibilities, role, and therefore status of an aunt. Baby Ruby successfully closes the signifying sequence with the statement in line 65: So your auntism: is: like nothing?. Judy and Baby Ruby signify by using the lexical, grammatical, prosodic, and interactional resources available to members of the African American community. Signifying in this interaction concerns how speakers assert and contest the unequal aunt/niece relationship and simultaneously negotiate the solidarity of age-based friendship. Judy and Baby Ruby recognize when they are the intended targets and verbally collaborate in signifying through a turn-for-turn matching of comparable resources. The skills they developed as children are used both to tease and to confirm, mediate, and constitute familial and personal relationships. Judy signifies that

No Woman No Cry

41

she is the aunt, but Baby Ruby signifies that Judy is much more her friend and peer and the "auntism is like nothing." Conclusion

This illustration of how African American girls, young women, and adults grow and function as core social actors in their communities demonstrates that they are part of rather than peripheral to vernacular culture. 10 Moreover, their identity as African Americans is tied to the construction of a cool social face, which they maintain and protect through language games such as he-said-she-said, instigating, and conversational signifying. When the great reggae artist Bob Marley sang "No Woman, No Cry," he included the chorus and mantra "Everything's gonna be alright." He was not singing about the mistakes of the past but about how we must take lessons from the past. In our case, these lessons lead us to reanalyze the research that has defined the field and to conduct new research that includes women as social actors. To understand the African American speech community and women's role in it is possible only when scholars are self-aware. We must realize what we say about others, as well as ourselves, when we privilege some groups and marginalixe others. And we must provide a sincere and exhausting critique when authors, analysts, and activists keep making the same mistakes. NOTES

Since the original writing of this chapter, two publications on similar topics have been published (Morgan 1996, 1998), but the theoretical arguments presented here are new. 1. I use African American English (AAE) to refer to the language varieties of U.S. residents of African descent. AAE acknowledges speakers' African descent and connects U.S. speakers with those in the African diaspora, especially the English-speaking diaspora in the Americas. American English (AE) refers to varieties of U.S. English without regard to social or cultural markedness or class, region, gender, or age. These varieties include standard, network, and mainstream, as well as working-class, Southern, Brooklyn, and so on. 2. African American coolness is similar to the Akan notion of dry speech as clear, precise, witty, and full of integrity. This contrasts with Akan wet or uncool speech, which is viewed as immature, dull, and slurred (cf. Yankah 1991a:47-54). See Morgan (1998) for a detailed discussion of this subject. 3. Such self-construction through narrative is found in other discourse communities as well; see the chapters by Lisa Capps, Marjorie Orellana, Patricia Sawin, and Kathleen Wood in this volume. 4. In the transcript, CAPITAL LETTERS indicate some form of emphasis, signaled by changes in pitch or amplitude. A period (.) indicates a fall in tone, not necessarily the end of a sentence, whereas a comma (,) indicates a continuing intonation, not necessarily between clauses of sentences. Colons (:) indicate that the preceding sound is lengthened. A question mark (?) indicates a rising intonation, not necessarily a question, and an exclamation point (!) indicates an animated tone, not necessarily an exclamation. Dashes and hyphens (-) indicate a cutoff or interruption of sound. Overlapping utterances are marked with a single left

42

IDENTITY AS INVENTION

bracket ([). A period within parentheses (.) indicates a one-second pause. Numbers in parentheses (2) are timed pauses in seconds. Double parentheses are transcriber comments. 5. This expression means 'confront' or 'fight'. 6. Although this is a reported story, I have other recordings of teenagers engaged in actual instigating. I have also helped mediate the preconfrontation stage of these episodes, but with only minor success. 7. This form of directed discourse is also called throwing shade. 8. Some people use the term reading to mean 'telling someone off, but in the absence of an audience this claim cannot be confirmed. 9. In recent examples of directed discourse in hip hop, artists "diss" opponents who fail to "come correct," "represent," or "give props." It is especially severe because audiences wait for artists to do battle. Such interactions can be found between artists Tim Dog and Dr. Dre, Dr. Dre and Eazy E, Kool Moe Dee and LL Cool J, and MC Lyte and Antoinette, among others. 10. Whereas African American women's discourse skills originate in their childhood language practices, Jennifer Coates (chapter 6, this volume), by contrast, finds a disjunction between younger white British girls' playful discourse style and their more disempowered style as they grow older. REFERENCES

Abrahams, Roger (1962). Playing the dozens. Journal of American Folklore 75:209—218. (1970). Deep down in the jungle. Chicago: Aldine. (1974). Black talking on the streets. In Richard Bauman & Joel Sherzer (eds.), Explorations in the ethnography of speaking. Cambridge: Cambridge University Press, 240-262. (1975). Negotiating respect: Patterns of presentation among black women. Journal of American Folklore 88:58-80. (1976). Talking black. Rowley, MA: Newbury House. Alleyne, Mervyn (1980). Comparative Afro-American: An historical-comparative study of Englishbased Afro-American dialects of the New World. Ann Arbor, MI: Karoma Press. (1989). Roots oj Jamaican culture. London: Pluto Press. Ball, Arnetha F. (1992). The discourse of power and solidarity: Language features of African American females and a male program leader in a neighborhood-based youth dance program. In Kira Hall, Mary Bucholtz, & Birch Moonwomon (eds.), Locating power: Proceedings of the Second Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 23-35. Butler, Judith (1995). Collected and fractured: Response to Identities. In Anthony Appiah & Henry Louis Gates Jr. (eds.), Identities. Chicago: University of Chicago Press, 439-447. Crenshaw, Kimberle (1992). Whose story is it anyway?: Feminist and antiracist appropriations of Anita Hill. In Toni Morrison (ed.), Race-ing justice, en-gendering power: Essays on Anita Hill and Clarence Thomas, and the construction of social reality. New York: Pantheon, 402— 440. Dollard, John (1973). The dozens: Dialectic of insult. In Alan Dundes (ed.), Mother wit from the laughing barrel. Jackson: University of Mississippi Press, 277-294. (Original work published 1939) Duneier, Mitchell (1992). Slim's table: Race, respectability, and masculinity. Chicago: University of Chicago Press. Duranti, Alessandro (1993). Truth and intentionality: An ethnographic critique. Cultural Anthropology 8(2):214-245. Eckert, Penelope, & Sally McConnell-Ginet (1992). Think practically and look locally: Language and gender as community-based practice. Annual Review of Anthropology 21:461-490.

No Woman No Cry

43

Etter-Lewis, Gwendolyn (1991). Standing up and speaking out: African American women's narrative legacy. Discourse and Society 2:425-437. (1993). My soul is my own: Oral narratives of African American women in the professions. London: Routledge. Etter-Lewis. Gwendolyn, & Michelc Foster (eds.) (1996). Unrelated kin: Race and gender in women's personal narratives. London: Routledge. Foster, Michele (1995). Are you with me?: Power and solidarity in the discourse of African American women. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routlcdgo, 329-350. Garner, Thurmon (1983). Playing the dozens: Folklore as strategics for living. Quarterly Journal of Speech 69:47-57. Gates, Henry Louis Jr. (1988). The signifying monkey: A theory of African-American literary criticism. Oxford: Oxford University Press. George, Nelson (1992). Buppies, b-boys, baps, and hohos: Notes on post-soul Black culture. New York: HarperCollins. Goffman, Erving (1967). Interaction ritual: Essays on face-to-face behavior. New York: Anchor Books. (1974). Frame analysis. New York: Harper & Row. Goodwin, Marjorie Harness (1980). He-said-she-said: Formal cultural procedures for the construction of a gossip dispute activity. American Ethnologist 7:674—695. (1982). "Instigating": Storytelling as a social process. American Ethnologist 9:76-96. (1985). The serious side of jump rope: Conversational practices and social organization in the frame of play. Journal of American Folklore 98:315—330. (1988). Cooperation and competition across girls' play activities. In Sue Fisher & Alexandra Dundas Todd (eds.), Gender and discourse: The power of talk. Norwood, NJ: Ablex, 55-94. (1990). He-said-she-said: Talk as social organization among black children. Bloomington: Indiana University Press. (1992). Orchestrating participation in events. In Kira Hall, Mary Bucholtz, & Birch Moonwomon (eds.), Locating power: Proceedings of the Second Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 182-196. Grier, William, & Price Cobbs (1968). Black rage. New York: Bantam. Gwaltney, John (1981). Drylongso: A self-portrait of black America. New York: Vintage. Hall, Stuart (1995). The question of cultural identity. In Stuart Hall, David Held, Don Hubert, & Kenneth W. Thompson (eds.), Modernity: An introduction to modern societies. Oxford: Blackwell, 596-631. Henley, Nancy M. (1995). Ethnicity and gender issues in language. In HopeLandrinc(cd.), Bringing cultural diversity into feminist psychology: Theory, research, practice. Washington, DC: American Psychological Association, 361-395. hooks, bell (1990). Yearning: Race, gender, and cultural politics. Boston: South End Press. (1992). Black looks: Race and representation. Boston: South End Press. Horton, John (1972). Time and cool people. In Thomas Kochman (ed.), Rappin' and stylin' out: Communication in urban black America. Urbana: University of Illinois Press, 19—31. Hurston, 7ora Neale (1990). Jonah's gourd vine. New York: Harper. (Original work published 1934) (1937). Their eyes were watching God. Philadelphia: Lippincott. Kochman, Thomas (1972). Toward an ethnography of Black American speech behavior. In Thomas Kochman (ed.), Rappin' and stylin' out: Communication in urban black America. Urbana: University of Illinois Press, 241-264. (1973). Review of Language behavior in a black urban community by Claudia MitchellKernan. Language 49(4):967-983.

44

IDENTITY AS INVENTION

(1983). The boundary between play and nonplay in Black verbal dueling. Language in Society 12(3):329-337. (1986). Strategic ambiguity in Black speech genres: Cross-cultural interference in participant-observation research. Text 6(2): 153-170. Kondo, Dorinnc (1997). About face: Performing race in fashion and theater. London: Routledge. Kunjuiu, Jawanza (1986). Countering the conspiracy to destroy Black boys. 2 vols. Chicago: African American Images. Labov, William (1972). Language in the inner city: Studies in the Black English Vernacular. Philadelphia: University of Pennsylvania Press. (1998). Co-existent systems in African-American Vernacular English. In Salikoko Mufwene, John Rickford, Guy Bailey, & John Baugh (eds.), African-American English: Structure, history, use. London: Routledge, 110-153. Major, Clarence (1994). Juha lo jive: A dictionary of African-American slang. New York: Penguin. Majors, Richard, & Janet Mancini Billson (1992). Cool pose: The dilemmas of black manhood in America. New York: Lexington. Mitchell-Kernan, Claudia (1971). Language behavior in a black urban community. Berkeley: University of California Language Behavior Laboratory. (1972). Signifying, loud-talking, and marking. In Thomas Kochman (ed.), Rappin' and stylin' out: Communication in urban black America. Urbana: University of Illinois Press, 315-335. (1973). Signifying. In Alan Dundes (ed.), Mother wit from the laughing barrel. Jackson: University of Mississippi Press, 310-328. Morgan, Marcyliena (1989). From down South to up South: The language behavior of three generations of Black women residing in Chicago. Ph.D. diss., University of Pennsylvania. (1991). Indirectness and interpretation in African American women's discourse. Pragmatics l(4):421-452. (1993). The Africanness of counterlanguage among Afro-Americans. In Salikoko Mufwene (ed.), Africanisms in Afro-American language varieties. Athens: University of Georgia Press, 423-435. (1994a). The African American speech community: Reality and sociolinguistics. In Marcyliena Morgan (ed.), The social construction of reality in Creole situations. Los Angeles: Center for African American Studies Press, 121—150. (1994b). Theories and politics in African American English. Annual Review of Anthropology 23:325-345. (1996). Conversational signifying: Grammar and indirectness among African American women. In Elinor Ochs, Emanuel Schegloff, & Sandra Thompson (eds.), Interaction and grammar. Cambridge: Cambridge University Press, 405-433. (1998). More than a mood or an attitude: Discourse and verbal genres in African American culture. In Salikoko Mufwene, John Rickford, Guy Bailey, & John Baugh (eds.), African-American English: Structure, history, use. London: Routledge, 251—281. Morrison, Toni (1992). Playing in the dark: Whiteness and the literary imagination. Cambridge: Harvard University Press. Mufwene, Salikoko (1992). Ideology and facts on African American English. Pragmatics 2(2): 141— 166. Percelay, James, Ivey Monteria, & Stephan Dweck (1994). Snaps. New York: Quill. Quirk, Randolph, et al. (1972). A grammar of contemporary English. London: Longman. Rickford, John (1986). The need for new approaches to social class analysis in sociolinguistics. Language and Communication 6(3):215—221. Rickford, John, Arnctha Ball, Rcnec Blake, Raina Jackson, & Nomi Martin (1991). Rappin on

No Woman No Cry

45

the copula coffin: Theoretical and methodological issues in the analysis of copula variation in African American vernacular. Language Variation and Change 3(1):103-132. Rickford, John, & Faye McNair-Knox (1993). Addressee- and topic-influenced style shift: A quantitative sociolinguistic study. In Douglas Biber & Edward Finegan (eds.), Sociolinguistic perspectives on register. Oxford: Oxford University Press, 235-276. Smitherman, Geneva (1977). Talkin and testifyin: The language of Black America. Boston: Hough ton Mifflin. (1994). Black talk: Words and phrases from the hood to the amen corner. New York: Houghton Mifflin. Spears, Arthur (1988). Black American English. In Johnnetta B. Cole (ed.), Anthropology for the nineties: Introductory readings. New York: Free Press, 96-113. Turner, Patricia (1993). 1 heard it through the grapevine: Rumor in African-American culture. Berkeley: University of California Press. Visweswaran, Kamala (1994). Fictions of feminist ethnography. Minneapolis: University of Minnesota Press. Washington, Mary Helen (1987). Invented lives: Narratives of black women 1860-1960. New York: Doubleday. Wolfram, Walter (1969). A sociolinguistic description of Detroit Negro speech. Washington, DC: Center for Applied Linguistics. Yankah, Kwesi (1991 a). Oratory in Akan society. Discourse and Society 2(l):47-64. (1991b). Power and the circuit of formal talk. Journal of Folklore Research 28(1): 1-22.

2

KATHLEEN M. WOOD

Coherent Identities amid Heterosexist Ideologies Deaf and Hearing Lesbian Coming-Out Stories

he last decade has seen a rise in antihomophobia legislation and a comercial production of positive images of queer lifestyles in the media, on prime-time television, and in Hollywood.1 Tempering this apparent acceptance is the reality that many queers still commonly experience both family rejection and public acts of hatred. In this social context, how and where do queers resist heterosexist boundaries and realize coherent identities as lesbians and gay men? One site is the life stories they tell, where identities are discursively created amid prevailing ideologies. Like other life stories about counterhegemonic living, the life stories of queers are oftentimes stories of a movement away from a social mainstream to life on the perimeter—stories of self-transformation about recognizing and revealing gender and attraction differences. These coming-out stories have been variously examined in the literature of anthropology and sociolinguistics (cf. Deby 1996; Gaudio 1996; Liang 1997; Wood 1997), providing us with a theoretical backdrop for the study of their structures. Julia Penelope and Susan Wolfe, in a widely read collection of lesbians' stories, explain that "coming out involves revealing one's identity to oneself, family members, colleagues, and communities. Coming out stories characterize the journey's two levels: the external events of becoming a lesbian, and the internal processes that accompany these events" (1989:10). In this chapter, I will examine the structures of coming-out stories of Deaf and hearing lesbians in order to show how coherent lesbian identities are realized, a narrative task that requires such women to contend with hegemonic notions of gender, attraction, and sexuality.2

Coherence and identity in life stories and personal narratives Social constructionists in anthropology, social psychology, rhetoric, and linguistics have discussed the impact of social structures such as gender, sexual orientation, race, 46

Coherent Identities amid Heterosexist Ideologies

47

class, linguistic background, and literacy ability on the discursive production of identity (e.g., Bakhtin 1981, 1986; Bruner 1990; Foucault 1982; Gergen 1992; Shotter 1993). This paradigm has little room for the rigid, monolithic view of identity taken by many linguistic variationists in which a specific linguistic form indexes a single identity category. Researchers from various fields have analyzed the discursive production of identities and cultures in autobiography (Bruner 1990; Rosen 1988), personal narrative (Labov 1972). oral history (Hymes 1981), language use (Kannapell 1989), and life stories (Linde 1993). Barbara Johnstone claims that narrative "structures our sense of self and our interactions with others, our sense of place and community" (1990:5). Others (Polanyi 1985; Polkinghorne 1991; Rosen 1988; Schank & Abelson 1995) support this tie between narrative and identity, showing that personal and social identity eannot be separated from stories of community and that stories cannot be separated from the social being. Yet little beyond the structural level is known of the narrative resources that create and display textual selves in social context. One resource for the narrative construction of selves is coherence. Researchers who integrate frame theory into their analyses of talk argue that local and societal context is central to the production and analysis of coherent discourse (cf. Ribeiro 1993, 1994; Schiffrin 1994; Tannen 1985). In examining how lesbians realize identities through stories, then, we must go below the surface to the narratives' underlying logic or coherence by examining their local (interactional) and global (ideological) contexts. The coherence of stories—that is, the logic invoked by their interactional and ideological context—is created by inferences, shared assumptions, and subsequent presuppositions that underlie the discourses of a community. In her discussion of the systems of coherence indexed in life stories, Charlotte Linde notes, "There is a limited number of coherence systems that can be present in a given culture at a given time, since one's addressee must at least recognize if not share any coherence system one chooses to use" (1993:165). Jerome Bruner (1990) likewise points out that personal narratives are simultaneously a recounting of significant events and a justification for life choices. Like all self-transformation stories, coming-out stories are rhetorical attempts to justify one's life, to realize acceptable selves by creating coherent identities. For a life story to be coherent, the justification of one's choices or experiences must be recognizable and acceptable. The comingout stories that follow are successful in that they access ideologies and associated narrative patterns that are common and familiar within a general heterosexist context, while invoking local, interactional forms.

Lesbians telling coming-out stories For five years, I have been a member of a national electronic-mail distribution list of a group of Deaf and hearing lesbian friends. We created this list in order to stay in touch with one another on a daily basis. Over the last two years, I have urged my friends, "Send me your coming-out stories." Although my original request was for electronic-mail stories, some women wrote them on paper or told them in American Sign Language (ASL).

48

IDENTITY AS INVENTION

My analysis of these stories supports what other researchers have suggested: Identity is not an immutable, monolithic realization. Following Elinor Ochs, I argue that the relation of language to identity is not "a straightforward mapping of linguistic form to social meaning"; rather, identity is "constituted and mediated by the relation of language to stances, social acts, social activities, and other social constructs" (Ochs 1992:336-337). Likewise, this analysis argues against essentializing the coming-out story or suggesting that it is very different in form from any other self-transformation story. Coming-out stories, like all self-transformation stories, reveal the tellers' identities as the result of their struggles with the ideologies surrounding their transformation. What distinguishes lesbian coming-out stories from other transformation stories is that heterosexist ideologies comprise the most salient pool of coherence resources from which a woman constructs herself as sexual, gendered, attractive, and attracted. Lesbian self-transformation stories are stories of resistance about women loving women amid sexism and homophobia, facts that both constrain and provide resources for the realization of lesbian identities. Such stories trace the transformation from lifestyles grounded in dominant ideologies to lifestyles that question these ideologies. In order for this narrative transformation to be coherent, narrative logic must adhere to the sociolinguistic conventions of other stories whose coherence systems are bound to dominant ideologies. Lesbians must therefore tell coming-out stories that index the pool of acceptable (that is, coherent) narrative selves, selves that are recognized as either conforming to or resisting heterosexist ideologies.

Interactiona and ideological resources and constraints Johnstone claims that "it is more enlightening to think of factors such as gender, ethnicity, and audience as resources that speakers use to create unique voices than as determinants of how they will talk" (1996:56). Johnstone's focus on the creativity of the individual, on circumstances as identity resources, is on target, but it also attributes an undue amount of volition to the user and obscures the fact that these resources are both bountiful and restrictive. For this reason I use the term resources and constraints rather than simply resources for such factors. Lesbian identities are created at the narrative intersection of dominant heterosexist ideologies and lesbians' counterhegemonic stories of gender, attraction, and sexuality. The narrative stances that lesbians take toward hegemony reveal their identities. By narrative stance, I refer to the intertextual relationship of the narrative self to the possible selves promoted by dominant ideologies. Heidi Hamilton calls for "a relatively more linguistic definition of intertextuality" (1996:64). She notes that Nikolaus Coupland has cautioned her to include in her analysis not only real-life experiences but also those that are "wholly fictional," for they likewise participate in the formation of stereotypes. In the spirit of Coupland's advice, I argue here for a more integrative view of intertextuality. I suggest that there are at least two levels of resources and constraints that function intertextually and contribute to the coherence of identities: (1) the local or interactional level that indexes, in a specific speech act, tellers' membership within a community of practice (Eckert & McConnell-Ginet 1992); and (2) the societal level, which includes ideolo-

Coherent Identities amid Hetemsexist Ideologies

49

gies and prior discourses, real or imagined, in which certain archetypes, stereotypes, and prototypes are invoked. Roger Schank and Robert Abelson describe narrative prototypes, which they call "skeletons," as follows: "Instead of telling all the details of a situation, we can index it as 'betrayal' or 'undermining my confidence' or 'ordering me around' or 'being inconsiderate.' Such prototypic stones need not be negative, of course. We have 'heroism' stories and 'defense of our nation' stories, and 'always there when I need you' stories as well. . . . Authors construct their own reality by finding the events that fit the skeleton convenient for them to believe" (1995:51-52). What are the underlying story-prototype resources that lesbians index when they tell their coming-out stories and realize their identities? What narrative "skeletons" do they invoke? How do these skeletons promote the coherent realization of identities? Again, narrative skeletons are not necessarily benign resources. In fact, for many lesbians, especially for some very butch lesbians, narrating a lesbian life can be a painful, unpleasant task of reconciling oneself with what one "should" be. A butch lesbian must come to terms with the way she looks, walks, talks, and dresses and must narrate a story with the omnipresent realization that mainstream society, in its mass-media acceptance of lesbians, would prefer that she live at the lipstick-lesbian end of the spectrum and both narrate and live a more palatable life. From this perspective, ideologies of gender, attractiveness, and attraction in mainstream U.S. culture are less resources than constraints that force a lesbian to resist or risk being pulled to the hegemonic, heterosexist center. Ideologies and cultural models provide life-story tellers with the tools to evoke the acceptable selves of a community of practice. In his examination of how fraternity men create powerful identities, Scott Kiesling (1996) considers how individual speakers create identities by invoking the models or roles that the fraternity's culture makes available. Michael Agar suggests in his discussion of "languaculture," however, that conversationalists do more than reproduce or conegotiate an immutable set of stereotypes, cultural models, or social facts. Once speakers have mastered the language and culture of a community of practice, they can begin to see languaculture as a set of resources (as opposed to constraints) through which they "can create, improvise, criticize, or struggle against, as you please" (Agar 1994:236). Likewise, although lesbians telling coming-out stories are constrained by the boundaries of narrative skeletons and acceptable selves (which index dominant and counterhegemonic ideologies), these constraints provide them with the opportunity to resist and to create themselves coherently yet uniquely. Analyzing lesbian coming-out stories When lesbians tell coming-out stories to a lesbian linguist who intends to make their stories public by analyzing their identity creation, they are engaging in a unique rhetorical occasion. In the nuances of this lesbian-to-lesbian interview occasion, we see how the local interactive situation (in this case a sociolinguistic interview) and the larger ideological context together provide the resources and constraints for each lesbian to realize her identity. The logic invoked in the interactive frames of the story and their accompanying schemata contribute to the story's coherence and point to

50

IDENTITY AS INVENTION

the underlying narrative archetypes, skeletons, presuppositions, and stances toward ideologies, which in turn reveal identities. Frames Deborah Tannen (1993) discusses "the power of expectation" in each of the levels of frames that come into play in narrative discourse and demonstrates how these expectations are revealed in surface-level linguistic features. In this study, the coming-out stories trigger at least four kinds of interactive frames that contribute to the making of coherent, recognizable, and acceptable lesbian identities. The first is what Tannen calls the storytelling frame, the rules that tellers invoke as they attempt to tell a story according to the social rules of that task. That is, in this culture, stories often have a beginning, a middle, and an end, with a central point. Tellers adhere to this frame or transgress it in some way, but the frame, embedded unconsciously in individuals and in society, is always invoked. As Tannen points out, "there are a number of ways in which subjects reveal that they have expectations about how to tell a story" (1993:25), as when the teller says she is having trouble knowing what to include in the story or how to end it (Wood 1994, 1997). It is quite common for lifestory tellers to have trouble closing the storytelling frame because the topic of a life story, a life still being lived, is not over. It is particularly difficult for a lesbian to close a story about coming out, for it is embedded in a heterosexist culture that forces her to deal with countless experiences of coming out, from those with intimates such as colleagues, friends, and families to those with strangers such as real estate agents and talkative airline seatmates. A second interactive frame is the performance frame. The difference between a story that is told versus one that is performed is that form is foregrounded in the latter. As Richard Bauman explains: All framing, . . . including performance, is accomplished through the employment of culturally eonventionalized mctacommunication. In empirical terms, this means that eaeh speech community will make use of a structured set of distinctive communicative means from among its resources in culturally conventionali/.ed and culture-specific ways to key the performance frame, such that all communication that takes place within that frame is to be understood as performance within that community. (1978:16)

He lists some of the "communicative means" that index a performed story: special codes, figurative language, parallelism, special paralinguistic features, special formulae, appeals to tradition, and disclaimers of performance. Some of the tellers on the e-mail list are professional storytellers, and others assume a tale-teller role on the list. These storytellers clearly invoke a performance frame when they write or sign their stories. Another frame is the addresser/addressee frame, in which certain forms index the addressee's discourse role or social identity. According to Isolda Carranza (1996), in this interactive phenomenon of dialogic text construction, the text shows traces of how it is continuously affected by the addressee's behavior. Texts are also shaped by the immediate communicative situation. For example, someone who tells

Coherent Identities amid Hetewsexist Ideologies

51

a life story in a crowded and noisy coffee shop may raise her voice or cut the story short because of the physical strain of the communication situation. Likewise, stories told on e-mail contain evidence that the teller is aware of an e-mail frame (a type of genre frame) and the protocol involved, ratified or not. Although little research addresses the telling of stories via electronic mail, Sara Kiesler, Jane Siegel, and Timothy McGuire (1984) suggest that social-psychological factors (time and information-processing pressures, absence of regulating feedback, lack of status and position cues, social anonymity, and computing norms and etiquette) influence these transmissions, forcing special forms to take the place of, for example, backchanneling cues. The last frame of expectation is the event frame, within which narrators describe real-world events and their role within them. Roy Baumeister and Leonard Newman note that "The ease of making a story is partly because there is an element of story narrative inherent in the events themselves as they happen. Stories are not just constructed after the fact as an aid to memory or explanation. Participants in events are aware of plans, goals, conflicts, links between actions, resolutions, and other narrative elements during (and sometimes even before) the episode" (1995:99). Other researchers refer to event frames as event scripts, narrative prototypes, or skeletons. These are the seemingly standardized versions of a real-life event indexed in surface features. For example, some lesbian coming-out stories index one of the following narrative skeletons: Woman is raised in a heterosexual environment. Woman behaves and looks like a heterosexual. Woman is exposed to a lesbian lifestyle. Woman switches to behaving like a homosexual. Woman is raised in heterosexual environment. Woman does not behave or look like a heterosexual. Woman experiences soeial pressure to conform. Woman continues living as she always has.

And regardless of structure, all lesbian coming-out narrative skeletons have an accompanying overlay of the real-world, often burdensome, event of self-disclosure. As Mary Elliott puts it, What [coming-out] narratives elide, or mention only briefly and then recoil from (as an issue and experience), is the act of giving away the secret, that terrifying crossing of the abyss. One after another, these narratives present a minute glimpse of the dread, panic, confusion, and uncertainty of the actual moment of disclosure and then, as if mimctically reproducing the performance of the coming-out act itself as an uncomfortable (always) or even shameful (sometimes) but necessary ordeal, move on as quickly as possible and without comment to lengthy pedagogical, ethical, and sociological defenses of the coming-out process. (1996:694)

52

IDENTTTY AS INVENTION

The event of coming out in a heterosexist society necessarily subverts ideologies of how we are to be women, how we are to be attractive and attracted, and how we are to behave sexually. By examining the various narrative frames in coming-out stories we can see how lesbians orient themselves to dominant ideologies, revealing their unique identities. Ideologies If identities are socially constructed via language and surrounding ideologies, then ideologies are the nucleus around which people orient themselves in order to reveal recognizable and acceptable identities. Gunther Kress and Robert Hodge define ideology as "a systematic body of ideas, organized from a particular point of view." They argue that "since normal perception works by constant feedback, the gap between the real world and the socially constructed world is constantly being reduced, so that what we do 'see' tends to become what we can say" (1993:5-6). What are the selves that dominant ideologies of gender, attraction, and sexuality allow us to "see" and say? Lesbian coming-out stories are inevitably embedded in heterosexist ideologies, which tellers manage both to invoke and to transgress (cf. Liang, chapter 15, this volume). As Dorothy Holland and Naomi Quinn observe, "Among alternative versions of what is legitimate and what is inevitable, a given ideology is most compelling if its Tightness engages the sense one has of one's own personal uprightness and worthiness, or if its inevitability engages the view one has of one's own inherent needs and capacities. These matters lie at the heart of our understanding of ourselves and our place in life" (1987:13). These all-pervasive abstract beliefs about what is right and legitimate, and thus most appropriate, pervade the discourses of the community; both mainstream and marginalized members of a society are constrained by dominant ideologies, and the identities of both groups are realized in relation to hegemony. That is, lesbians are able to see themselves through the same ideological lens through which homophobes see them. While working on this chapter, I discovered that the homosexual "recovery group" Exodus was meeting on my campus for a three-day workshop, "Healing Homosexuality." The materials Exodus distributed conveniently indexed several heterosexist ideologies, which emanated from an overarching ideology that although homosexuality may be an inborn orientation, to act on it (that is, to have sexual relationships with members of the same sex) is a sexual sin. In one brochure Sy Rogers of the Church of Our Savior, a major ministry of homosexual conversion, discusses whether homosexuality is a sin or a gift: "As for Jesus, he did not specifically mention homosexuality. But then, he never mentioned incest, rape or bestiality for that matter. . . . Jesus showed mercy to those guilty of violating moral law—such as the woman caught in adultery. Yet He also commanded her to obey God with this second chance, and leave her life of sexual sin." Although the Exodus group represents a small segment of the heterosexist population, it clearly indexes mainstream ideologies. Another useful source of information about heterosexist ideologies is the media, as shown by Shari Kendall's (1996) study of news stories concerning Salt

Coherent Identities amid Heterosexist Ideologies

53

Lake City's 1996 ban of student gay groups in high schools. Kendall suggests that the heterosexist ideologies she locates emanate from the overarching ideology Homosexuality is not natural and "not natural" is not good:3 1. Homosexuality can be promoted. 2. Homosexuality is a decision. 3. There is no such thing as a homosexual, just an adoption of behaviors that make someone different. 4. Gay men and lesbians are rebellious children who don't know any better. 5. A homosexual lifestyle is different from a nonhomosexual lifestyle. 6. Homosexuality is a choice. 7. Homosexuality is not innate. 8. Homosexuality is sick. 9. All homosexuals are men.

For example, Kendall found that the textual presupposition that queers are less than human (that is, "unnatural") underlies the following excerpt from the Salt Lake Tribune (Feb. 3, 1996): As for [State Senator Charles Stewart], he says he has gotten nothing but support for his stance. While he has no anti-gay bills planned this session, he thinks other legislators might. And if some Utahns don't agree with him, Stewart makes no apologies. "It is a divisive issue for the whole society," he says. "It is drawing a line in the sand of what is civil and what is bestial. What is a human being and what is an animal."

Of course, one can argue that the articles and op-ed pieces Kendall examined are atypical of the U.S. mainstream and that this homophobic ideology is not culturally dominant. However, if this senator enjoys, at the very least, the support of those who elected him, and if this support enables him to proceed along a hegemonic trajectory, receiving privileged access to the media and an opportunity to declare what it means to be normal in this society, then Kendall's list of presuppositions does indeed represent a common sentiment within a heterosexist environment. In such an environment, gay men's and lesbians' lives are sites of hegemonic resistance. So too are their life stories and the identities they create in telling them. In fact, there would be no need for a "coming-out" story in a society where queer lifestyles enjoyed the mainstream support that heterosexuality does in the United States today. The tellers of these stories therefore narrate from a vantage point of resistance, which allows them to access possible selves beyond the confines of heterosexual society. Coming-out stories and the invention of identities I collected the data for this study between 1994 and 1996. Table 2.1 is a summary of the profiles of the participants and the form in which I received their stories. Each woman did what all life-story tellers do: She created identity through frames amid surrounding

54

IDENTITY AS INVENTION

Table 2.1. Study participants and story formats, by communication mode and language choice Storyteller

Deaf or hearing

Communication mode

language used

Cindy

Deaf

E-mail

English

Barbara

Deaf

E-mail

English

Kim

Hearing

E-mail

English

Maria

Deaf

E-mail

English

Ellen

Deaf

Videotaped

ASL

Jennifer

Deaf

Handwritten

English

ideologies. One of the most important tasks the tellers accomplish is to establish their place within a community of practice, in this case, a geographically dispersed group of Deaf and hearing lesbians. Another important task is to orient themselves to dominant heterosexist ideologies, an act which is itself the realization of identity. Cindy's e-mail story On the day I sent my e-mail message requesting coming-out stories to the list, Cindy quickly responded with two messages constituting her fourteen-line coming-out story.4 Cindy told me about her gay uncle who had had a nervous breakdown and had come out to his sister, Cindy's mother. He told Cindy that her mother had already guessed that Cindy too was gay. Cindy then delivered "all kind of book and information" to her mother under the guise of supporting her uncle, but eventually she came out to her. At this point in the narrative, Cindy tells me, in an embedded coming-out story, what she told her mother. It is the talc of her marriage, four children, grandchildren, and divorce. She ends by reporting that she is with her second and permanent "woman," "for good." Cindy narrates within an e-mail frame, invoking the protocol of the list. The following unratified rules have emerged on the list over time: 1. 2. 3. 4. 5. 6.

Be brief. Limit posting to one screen. Send without editing. Leave grammatical and mechanical errors in. Don't capitalize if you don't want to. Use "TTY" forms. Use ASL-English contact forms. Use humor whenever possible.

Cindy's orientation to the e-mail frame can be seen in her conformity to these rules. For example, she uses / six times in two lines, capitalizing it only twice. She uses woman instead of women for the plural form at one point and does not adhere to the subject/verb agreement rules of Standard English. These forms could be secondlanguage errors, since Cindy is a native speaker of ASL, or the result of quick, unedited

Coherent Identities amid Heterosexist Ideologies

55

typing. In any case, they contribute to the realization of Cindy's lesbian identity on e-mail: Her adherence to the list protocol aligns her with the lesbian members in this community of practice and indexes a larger disregard for the conventions of standard written English. Cindy, like many Deaf people who communicate on the phone with a teletype communication device (TTY), has transferred TTY conventions to electronic-mail communication. For example, she twice uses the abbreviation abt for about. Cindy thereby marks herself both as a member of the list, which expects such stylistic transfer, and as a regular TTY user and thus part of the larger Deaf community. Other aspects of the e-mail frame in Cindy's story relate not to the protocol for posting to the list but to the limitations of the mode of communication itself. For example, Me dont know how to tell my story abt my coming out hmm it started with my uncle who is gay and have been for years. The form hmm is a conversational placeholding marker, something that would likely occur in a face-to-face conversation. Here it substitutes for the intimacy of unmediated interaction. Cindy also invokes the storytelling frame: Her statement me dont know how to tell my story indicates her awareness of the special demands of story construction. Cindy thus uses sociolinguistic resources provided by e-mail and storytelling frames. But how is Cindy-as-lesbian realized in relation to dominant ideologies? She begins her story with, My mom ask me why i became gay I told her i was crazy abt women ever since i was a kids but thought I can t have that so i got marry and had four great kids and grandchildren. Cindy's mother's question indexes a presupposition that Cindy was not born gay, that her homosexuality is not innate (item 7 on the above list of homophobic ideologies). But Cindy's initial response indexes a presupposition that homosexuality is innate, or at least can occur in very early childhood. The immediacy of Cindy's response points to her current stance toward her sexuality and her orientation to dominant ideologies. Cindy tells her mother, "but i thought I can t have that so i got marry . . . ," a comment that indexes the presupposition that lesbianism is a set of behaviors (item 3 in the preceding list). Cindy in the narrative assumed she could "choose" to adopt heterosexuality if she wanted to, by living like a straight woman. Cindy the narrator hence draws on several narrative presuppositions and engages in a dialogue with the underlying ideologies, thereby indexing a particular narrative skeleton. The skeletons I present here are intended to represent not summaries of the coming-out stories but rather the underlying coherence model that the tellers apparently invoke in order to "make sense of their own lives" (Schank & Abelson 1995:59). Indexing established cultural models or narrative skeletons is an ongoing part of realizing a lesbian identity. "Baby dykes" (women new to lesbian communities of practice) may be more uncomfortable telling coming-out stories because they have not learned the skeletons for creating acceptable selves. (The issue of socialization into queer communities of practice is explored in greater detail by William Leap [chapter 13, this volume].) Cindy's coming-out story skeleton is as follows: Lesbian is raised as heterosexual. Lesbian feels attracted to women.

56

IDENTITY AS I N V E N T I O N

Lesbian decides to ignore lesbian attractions and behave as a heterosexual. Lesbian acknowledges being attracted to women and lives as a lesbian.

The above skeleton represents a model for the life events indexed by Cindy's coming-out story. But it is through Cindy's narrative stance to these events, her orientation to the underlying ideologies, that we come to know Cindy as a lesbian we can recognize and accept. The trajectory of her coming out is familiar and coherent. By invoking the e-mail and storytelling frames, Cindy is able to maneuver around at least two underlying ideologies: (1) homosexuality is not innate, and (2) homosexuality is a set of behaviors. She indexes a skeleton in which her identity as a lesbian clearly transgresses these dominant heterosexist ideologies. This transgression reveals Cindy as lesbian.

Barbara's e-mail story Barbara, also Deaf, sent a twenty-thrcc-line story which chronicles her sense that she was always a lesbian even when she attempted to date boys in high school. Barbara's narrative skeleton is identical to Cindy's. Queers and heterosexuals alike can recognize the narrative self in such a skeleton because it represents an acceptable coherence pattern for indexing the narrative trajectory, the movement of the self within the story world. Also like Cindy, Barbara uses several linguistic forms that indicate her adherence to list protocol. Her story is shorter than many of the others, and although she uses capital letters, she does not edit for spelling and grammar. She also follows one of the unratified list rules ("Use humor"): but never felt so comfortable always panorid at the end of dating with men like do i have to kiss, then petting or necking ugh!!!!! haha. Additionally, her use of the paralinguistic marker haha, which signals intended humor, reflects the rule "Use 'TTY' forms," for paralinguistic markers are very common in TTY communications. Barbara's story also conforms to the list rule to use ASL-English contact forms. This raises wider narrative questions: If linguistic forms contribute to the coherence of this story, what happens if the life-story teller is bilingual? And how are contact forms used to realize coherent identities? These data suggest that bilinguals have a wider range of language-form resources with which to signify identity. Barbara employs a type of linguistic borrowing that is common in TTY and e-mail conversations among ASL-English bilinguals: she uses wow to portray her elation at the large number of women on campus when she arrived as a new freshman (/ always prefer to be hanging around girls that time so when I came to Gallaudet wow more women there ha and still hanging around girls too). The English exclamation form wow is often lexicalized in ASL, fingerspelled rapidly in the air W-O-W, usually with both hands near the forehead. This form was perhaps first borrowed into ASL, then lexicalized (becoming a single sign), then borrowed back into English via TTY and e-mail communications, as in Barbara's story. She uses a similar form a little later in her narrative: I did try to be hanging arund men to get married and have Idd.s.... eeekkkk no way. The form eeekkkk here is also lexicalized, signed

Coherent Identities amid Heterosexist Ideologies

57

with one hand holding the sign for E near the mouth and forcefully moving away into the sign for K. These features are embedded in the narrative skeleton. Barbara indexes heterosexist ideologies when she explains how she tried to conform to heterosexual dating rituals: "and did try go out with boys but never felt so comfortable always panorid at the end of dating with men like do I have to kiss, then petting or necking ugh!!!!!! haha solhave to get lots of drink to ignore my feelings formento relax . . . . no fun!!!!!! ha so for society's sake I did try to be hanging arund men to get married and have kids . . . . eeekkkk no way." Here again we see the ideology located by Kendall and indexed by Cindy: the belief that if one acts heterosexual, one can, in fact, be heterosexual. Barbara evokes—and rejects—this ideology in asserting that she tried to act straight, but it didn't work. Barbara thus manages a coherent lesbian identity, recognizable because of her orientation to dominant ideologies. We recognize Barbara as lesbian (that is, her story is coherent) because she indexes a larger, almost commonsensical theme that if one tries something that doesn't work out, then one should move on. Kim's e-mail story Kim sent the most distinctive e-mail story, a three-line narrative embedded in a longer message. After explaining that she was afraid to risk composing the long version of her coming-out story because of technical problems with her computer, she wrote: so i'll give you the short version: girl meets girl, girl falls in love with girl, girlz ride into the sunset to (town name), girlz live happily ever after.

In composing her coming-out story as a synopsis, Kim invokes the storytelling frame, specifically the love-story archetype. But by using this frame and then substituting girl in the place of boy, Kim transgresses the archetype, which draws on a heterosexist ideology present in fairy tales: Girls meet boys and live happily ever after (cf. Orellana, chapter 3, this volume). Her story works because, as Linde (1993) suggests, we recognize (but do not share) the system of coherence it invokes. This transgressed archetype produces Kim's identity as a lesbian in a heterosexist society. Her narrative skeleton is a slight variation of Cindy's and Barbara's: Lesbian meets lesbian and falls in love. Lesbian lives as a lesbian.

Maria's e-mail story Maria's story bears very little resemblance to Cindy' s, Barbara's, or Kim's, except that it shares the same skeleton as Cindy's and Barbara's. Maria narrates that she always knew she was attracted to women and ended up entering a Roman Catholic convent, where she met the first woman she ever fell in love with. Her story is a description of

58

IDENTITY AS I N V E N T I O N

her family and the women she has loved. Maria's story is clearly more written than the others, which may be a result of Maria's identity as a professional writer. Although Maria is a member of the distribution list, she does not adhere to its rules. In fact, the only rule she follows is the elimination of capital letters; she uses not a single one in her ninety-line story. Moreover, she begins her story without either invoking a storytelling frame or expressing difficulty with the task (such as / don't know where to start or I'm not sure what you want): "i* ve known since i was old enough to know my own name that i like girls especially, it's just the way i am. and growing up in north dakota in the 40s and 50s, there was never any mention of words like 'lesbian.' i was well into college before i ever ran across the notion." She narrates her days in the convent, her departure, her marriage and eventual divorce, and her relationships with nuns still in the convent. The unusual structure of her story is also evident in the words she uses when she describes one such relationship: "during all of this time, i'd ask her periodically to leave the convent and live with me. and she would always refuse, after a while, it began to rankle, her second crop of novices arrived, and among them was a wonderful young woman with whom i promptly, though much against my better judgment, fell in love, she left to go back home and finish her masters, and i was bereft." This formal register—constituted in part by words like rankle and bereft—is highly marked on the list, indicating that Maria may not be specifically invoking the e-mail frame and is instead marking her text as a performance. Maria's lesbian identity is realized as she indirectly addresses ideologies of the origin of lesbianism: I've known since i was old enough to know my own name that i liked girls especially, it's just the way i am. Maria's use of the adverbial absolutive particle just responds to something that was never asked, as if someone has asked Maria to explain why she is a lesbian or how she became one. This may be the relationship between all personal narrative and public ideologies: responding to that which may not have been uttered but is all-pervasive. Maria addresses (actually transgresses) the ideology Homosexuality is not innate but does not make overt reference to any other ideologies and does not argue for her identity, implying that \hejust should end that part of the dialogue. Maria creates her identity in her coming-out story by evoking a recognizable lesbian narrative skeleton and by enlisting formal linguistic forms, which mark her as a writer-lesbian. Ellen's ASL story A similar invocation of these two identities—lesbian and professional teller of tales— is achieved very differently when different linguistic resources are available. Ellen, a Deaf lesbian who grew up in a large urban community, is a professional storyteller. In a videotaped interview with me, she narrated several events in her life, including early adolescence when she was oftentimes mistaken for a boy. She explains in one narrative sequence thai when she was about 12 years old a screaming woman who mistook her for a boy pushed her out of a women's rest room. A policeman rushed up to Ellen, questioned her, groped her crotch, and demanded that she pull down her pants to prove she was a girl. How did Ellen coherently realize this frightened-youngbutch identity?

Coherent Identities amid Heterosexist Ideologies

59

Because ASL is a visuospatial language, the teller can allow the audience to shape the text. The active role of the audience is central to the effectiveness of Ellen's narrative. In ASL, the narrator has at least two choices for representing the complicating action of a story. She may narratively distance herself by using a placeholder or classifier sign for herself to show the action that happened to the protagonist of the story (herself). The signing space of this type of narrative is removed, literally occurring further out from the body with narrative/spatial trajectories emanating from this point in the signing space. Or she may take on the role of her past self (and the characters in her story), imitating countenance, body position, and action. Ellen used the second technique for this part of the story. First, in a form of direct discourse, Ellen "became" her young self, a lesbian who was publicly harassed because of her gender-bent appearance. In this same narrative, through the use of role shifting (Padden 1986), another common feature of ASL narrative, Ellen also became the policeman who groped her. And because I was the audience with whom she maintained eye contact, when she "was" the policeman, the direction of the policeman's grope was toward me. In a sense, I briefly became "Ellen the young dyke" and felt the violation of that grope. I could see Ellen at that age in a way that no spoken version would have given me access to. A few minutes later, she explained that in order to avoid harassment, she began using the men's rest room instead. The performative aspects of Ellen's story recall Maria's creatively written narrative. Ellen clearly invoked a performance frame, setting up the camera and handling the remote control. She began her story by emerging from a coat closet and looking directly into the camera, signing, "Should I come out now?" Splicing her story by turning the camera on and off, she taped herself going back into the closet before emerging again and beginning her story. After nearly an hour and a half of taping, and after her story was clearly finished and we had moved on to other topics, she jumped up, explaining that she had to "close the story." She positioned herself near the closet, turned on the camera, and walked backward into the closet from where she began the story, thus providing parallel closure. Here she invoked the performance frame, something she is accustomed to doing in her professional life, and created a lesbian identity by using the figurative language of the expression coming out of the closet. Thus far I have focused on how the storytellers deal with prevalent ideologies about attraction. In this section of Ellen's coming-out story, however, the ideologies she confronts concern appearance: She was harassed because she did not look like a girl. We can presume the narrative coda, from then on, I used the men's room, to be a statement about her coming of age in a heterosexist society that dictates how gender should be realized. Again, this coda represents Ellen's transgression of hegemonic notions and allows us to understand her lesbian identity. Ellen's is the only comingout story 1 received that embodies the following "butch" skeleton, but it is a common storyline: Lesbian is raised as heterosexual. Lesbian rebels early, becoming a butch. Lesbian acknowledges her attraction to women.

60

IDENTITY AS INVENTION

Lesbian faces social disapproval. Lesbian continues to live a butch existence.

Jennifer's handwritten English story In her exploration of the intersection of Deaf and lesbian identities in the life story of a 28-year-old lesbian, Tina Neumann (1997) found that the coming-out-as-a-lesbian process was similar to the coming-out-as-Deaf process. Because of heterosexism, most lesbians go through a self-transformation process from a heterosexual childhood into a lesbian community or lesbian consciousness. Likewise, because approximately 90 percent of Deaf people are born into hearing families, realizing a Deaf identity requires a parallel self-transformation—from a hearing-based childhood and family to a Deaf-centered one. In Deaf lesbians' coming-out stories, it is not uncommon for both event frames, coming out as Deaf and coming out as lesbian, to be invoked and to create narrative coherence. We can see this dual "coming out" in Jennifer's coming-out story. Jennifer begins with a narrative about realizing, after several dates with her male high school peers, that she was a lesbian. She also explains coming to terms with "becoming deaf in '69 and then Deaf in '77." She describes how she manages this intersection of "comings out": "Long before 1984 I knew I liked the company of womon. When I double-dated w/ my boyfriends I was always more interested in what the womon had to say. (They also tried to enunciate a lot clearer & that helped w/ lipreading. Sure I knew signs but all my boyfriends except 2 were non-Deaf and signing-impaired. But that's another story for another time.)" In these lines, Jennifer invokes both event frames, coming out as lesbian and as Deaf. Like Maria's e-mail story, Jennifer's story does not make use of any of the typical e-mail frame devices. Her story has no TTY abbreviations and no ASL or contact forms, nor does she invoke an interactive frame. Her story has no title and no reference to me, the person who asked her to write it. Instead, she begins the story as follows: "When I'm asked how or when did I 'come out' I get this reaction where I freeze and there is this panorama of dates and places and people. Each and every one of those dates and places and people I spoke to about my sexual & cultural identity as a 'New Dyke on the Block' are equally very much a part of the 'process' of my coming out." Jennifer, like Maria, realizes her identity through the use of a more formal register, invoking special forms such as New Dyke on the Block and womon, and / was by the flood of newly sensual emotions aroused by this later in the narrative. In addition to the English forms she uses to realize her Deaf and lesbian identities, Jennifer also invokes dominant heterosexist ideologies. In a later section, she explains that she wrote coming-out letters to family members and got varying responses: At one other point I got a letter from my aunt telling me that my being a lesbian was "just a stage" in my life?! The final punctuation is in boldface, which indicates a textual dialogue with the heterosexist ideology that Jennifer's aunt invoked, as if Jennifer were saying to the audience, "What?! Yes, she really did say that!" Although Kendall (1996) does not mention the ideology of homosexuality as "just a phase," it is a common assumption, related to the ideology that "gay men and lesbians are rebellious children who don't know any better" (item 4 in the preceding list). Jennifer creates herself as lesbian by narratively butting up against this assumption and covertly transgressing it.

Coherent Identities amid Helerosexist Ideologies

61

Jennifer's coming-out story, embedded in the political peace movements of the last few decades, shows how life stories can have both major and minor narrative skeletons. It also represents how women of minority groups necessarily deal with a kind of coming out concerning their status in the dominant culture. Jennifer tells a coming-out story that includes every facet of her identity: She is Deaf, she is politically active, and she is a lesbian. In this way, Jennifer reveals an identity that is linked to her acceptance of her deafness and to her political coming of age during the late 1970s and early 1980s. Her story invokes a not-uncommon coming-out skeleton: Lesbian is raised as heterosexual and hearing. Lesbian is attracted to girls. Lesbian accepts her deafness and lives in the Deaf-World. Lesbian acknowledges her attraction to women. Lesbian lives as a lesbian.

Conclusion I have shown how women in a single community draw on local and social systems of coherence to create their identities in coming-out stories. This analysis suggests that identity is both locally situated and embedded within the larger ideological context. As Deborah Schiffrin puts it, "who we are is, at least partially, a product of where we are and who we are with, both in interactional and story worlds" (1996:198). I have shown how frames are used to appeal to locally constructed norms for realizing identity. We have seen how lesbians' coming-out stories invoke certain narrative skeletons, which reveal the lesbian-coming-out-story schemata and prototypes. And finally, these data have shown how identities are necessarily created within the ideological context in which they exist. Paradoxically, although lesbian identities exist outside of heterosexist ideologies, it is by means of such ideologies that they are recognizable and acceptable. In order to realize identities, lesbians at once draw upon and transgress ideologies as resources for coherence. NOTES

I am grateful to Shari E. Kendall and the students in her course "Women, Men, and Language" at Georgetown University (spring 1997). In their analysis of heterosexual undergraduates' dating-and-attraction stories, they concluded that heterosexuals also construct gendered identities that support or resist hegemonic notions of what it means to be a woman or man and what it means to be attractive and attracted. Their comments steered me to the conclusions I make here about lesbians' coming-out stories, which, I would argue, are the queer analogue to the Georgetown students' dating stories. Thanks also to Jules NelsonHill, Carol Monigan, Kendra Smith, MJ Bienvenu, Shari Kendall, Keller Magenau, Galey Modan, and E. Lynn Jacobowitz. 1. I use the term queer to include lesbians, gay men, bisexuals, and transsexuals. 2. Deaf (capital D)—as opposed to deaf (lowercase d)—represents those deaf people politically and culturally aligned with the "Deaf-World." The term "Deaf-World" is recog-

62

IDENTITY AS INVENTION

nized by Harlan Lane, Robert Hoffmeister, and Ben Bahan (1996), who explain that it is the name members of the ASL-signing Deaf community call themselves. 3. It is interesting to note that the homophobic logic surrounding the arguments against gays and lesbians is strikingly similar to that of the Exodus group literature. 4. Due to length restrictions, full transcripts of the stories are not provided. 5. See note 2. REFERENCES

Agar, Michael (1994). Language shock: Understanding the culture of conversation. New York: Morrow. Bakhtin, Mikhail M. (1981). The dialogic imagination. Ed. Michael Holquist. Trans. Caryl Emerson & Michael Holquist. Austin: University of Texas Press. (1986). Speech genres and other late essays. Ed. Caryl Emerson & Michael Holquist. Trans. Vern W. McGee. Austin: University of Texas Press. Bauman, Richard (1978). Verbal art as performance. Rowley, MA: Newbury House. Baumeister, Roy F., & Leonard S. Newman (1995). The primacy of stories, the primacy of roles, and the polarizing effects of interpretive motives: Some propositions about narratives. In Robert S. Wyer, Jr. (cd.), Knowledge and memory: The real story. Hillsdalc, NJ: Erlbaum, 97-108. Bruncr, Jerome (1990). Acts of meaning. Cambridge, MA: Harvard University Press. Carranza, Isolda (1996). Argumentation and ideological outlook in storytelling. Ph.D. diss., Georgetown University. Deby, Jeff (1996). Looking into coming out: Why tell a coming-out story? Paper presented at the annual meeting of the Georgetown Linguistics Society, Washington, DC. Eckert, Penelope, & Sally McConnclI-Ginet (1992). Communities of practice: Where language, gender, and power all live. In Kira Hall, Mary Bucholtz, & Birch Moonwomon (eds.), Locating power: Proceedings of the Second Berkeley Women and Language. Conference. Berkeley, CA: Berkeley Women and Language Group, 89-99. Elliott, Mary (1996). Coming out in the classroom: A return to the hard place. College English 58(6):693-708. Foucault, Michel (1982). The subject and power. Critical Inquiry 8:777-795. Gaudio, Rudolf P. (1996). Out in the open without coming out: Queer narratives from Nigerian Hausaland. Paper presented at the annual meeting of the Georgetown Linguistics Society, Washington, DC. Gergen, Mary (1992). Life stories: Pieces of a dream. In George Rosenwald & Richard Ochberg (eds.), Storied lives. New Haven, CT: Yale University Press, 127-144. Hamilton, Heidi (1996). Intratextuality, intertextuality, and the construction of identity as patient in Alzheimer's disease. Text 16(1):61-90. Holland, Dorothy, & Naomi Quinn (1987). Culture and cognition. In Holland & Quinn (eds.), Cultural models in language and thought. New York: Cambridge University Press, 3-40. Hymes, Dell (1981). "In vain I tried to tell you": Essays in Native American ethnopoetics. Philadelphia: University of Pennsylvania Press. Johnstone, Barbara (1990). Stories, community, and place: Narratives from Middle America. Bloomington: Indiana University Press. (1996). The linguistic individual. New York: Oxford University Press. Kannapell, Barbara (1989). An examination of Deaf college students' attitudes toward ASL and English. In Ceil Lucas (ed.), The sociolingui sties of the Deaf community. New York: Academic Press, 191-210. Kendall, Shari E. (1996). Conflicting ideologies in the news coverage of Salt Lake City's ban on

Coherent Identities amid Heterosexist Ideologies

63

gay and lesbian clubs in public schools. Paper presented at the annual meeting of the Georgetown Linguistics Society, Washington, DC. Kiesler, Sara, Jane Siegel, & Timothy W. McGuire (1984). Social psychological aspects of computer-mediated communication. American Psychologist 39(10):! 123-1134. Kiesling, Scott (1996). Language, gender, and power in fraternity men's discourse. Ph.D. diss., Georgetown University. Kress, Gunther, & Robert Hodge (1993). Language as ideology. New York: Routledge. Labov, William (1972). Language in the inner city. Philadelphia: University of Pennsylvania Press. Lane, Harlan, Robert Hoffmeister, & Ben Bahan (1996). A journey into the Deaf-World. San Diego: Dawn Sign Press. Liang, A. C. (1997). Creating coherence in coming-out stories. In Anna Livia & Kira Hall (eds.), Queerly phrased. New York: Oxford University Press, 287-309. Linde, Charlotte (1993). Life stories: The creation of coherence. New York: Oxford University Press. Neumann, Tina (1997). Deaf identity, lesbian identity: Intersections in a life narrative. In Anna Livia & Kira Hall (eds.), Queerly phrased. New York: Oxford University Press, 274-286. Ochs, Elinor (1992). Indexing gender. In Alessandro Duranti & Charles Goodwin (eds.), Rethinking context. New York: Cambridge University Press, 335-358. Padden, Carol (1986). Verbs and role-shifting in ASL. In Carol Padden (ed.), Proceedings of the Fourth National Symposium on Sign Language Research and Teaching. Silver Spring, MD: National Association of the Deaf, 44-57. Penelope, Julia, & Susan Wolfe (1989). The original coming out stories. Freedom, CA: Crossing Press. Polanyi, Livia (1985). Telling the American story: A structural and cultural analysis of conversational storytelling. Norwood, NJ: Ablex. Polkinghorne, Donald E. (1991). Narrative and self-concept. Journal of Narrative and Life History 1(2/3):135-153. Ribeiro, Bianca Telles (1993). Framing in psychotic discourse. In Deborah Tannen (ed.), Framing in discourse. New York: Oxford University Press, 77-112. (1994). Coherence in psychotic discourse. New York: Oxford University Press. Rosen, Harold (1988). The autobiographical impulse. In Deborah Tannen (ed.), Linguistics in context: Connecting observation and understanding. Norwood, NJ: Ablex, 69-88. Schiffrin, Deborah (1994). Approaches to discourse. Cambridge: Blackwell. (1996). Narrative as self-portrait: Sociolinguistic constructions of identity. Language in Society 25:167-203. Schank, Roger C., & Robert P. Abelson (1995). Knowledge and memory: The real story. In Robert S. Wyer (ed.), Knowledge and memory: The real story. Hillsdale, NJ: Erlbaum, 1-86. Shotter, John (1993). Conversational realities. London: Sage. Tannen, Deborah (1985). Frames and schemas in interaction. Quaderni di Semantica 6:326-335. (1993). What's in a frame? Surface evidence for underlying expectations. In Deborah Tannen (ed.), Framing in discourse. New York: Oxford University Press, 14-56. Wood, Kathleen (1994). Life stories as artifacts of a culture: Lesbian coming-out stories. In Mary Bucholtz, A. C. Liang, Laurel A. Sutton, & Caitlin Mines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 777-786. (1997). Iconicity in lesbian coming-out stories. In Anna Livia & Kira Hall (eds.), Queerly phrased. New York: Oxford University Press, 257-273.

3

MARJORIE FAUL5TICH O R E L L A N A

Good Guys and "Bad" Girls Identity Construction by Latina and Latino Student Writers

lassroom writing workshops—spaces created within school environments for students to write, revise, and "publish" texts—have received considerable attention from literacy researchers over the last decade (see, for example, Atwell 1987; Calkins 1986; Graves 1983). Yet only a few researchers have explored the social negotiations that take place within free writing environments (Dyson 1989,1993; Lensmire 1994; Phinney 1994) and their gendered dimensions, where stereotypes may be reproduced, reinforced, and implicitly sanctioned in children's texts (Gilbert 1993; Kamler 1994; MacGillivray & Martinez 1998). With few exceptions (e.g., Brooke 1991), most studies of school writing processes focus on cognitive processes and give little attention to students' uses of language to explore selves or construct social identities. The children I observed in a writing workshop environment during two overlapping school years were involved in much more than the acquisition of literacy. In this chapter, I examine how these Latina and Latino primary-school children experimented with a range of social identities and made sense of what it means to be a girl or a boy, rich or poor, brown or white, good, bad, beautiful, or smart, through the characters they created in their stories. 1 view language and literacy as important tools that the children used for the invention of possible selves and examine the written products—more than 300 stories produced over a 13-month period—for the insights they provide into the inventors.

Inventing identities through language and literacy A growing body of work, to which writers in this volume contribute, examines the construction of identity through language practices and the linguistic presentation of 64

Good Guys and "Bad" Girls

65

self as a member of particular social groups. Language is examined as an integral part of "doing" membership in these groups, something continually invented and reinvented in interaction with others (Barrett, chapter 16, this volume; Cook-Gumperz 1995; Gal 1995). This new approach emphasizes human agency, highlighting choices that people make in adopting particular language forms and assuming membership in discourse communities. Yet researchers recognize that the freedom to invent identities is constrained by social forces. Bronwyn Davies (1993) explores the gendered discourses that frame children's talk about texts that they read and write, including variations in the construction of "femininity" in the talk of girls from different socioeconomic groups. Jennifer Coates (chapter 6, this volume) examines the range of discourses that a group of white middle-class teenage girls take up; through their talk they "do" both friendship and femininity. Lisa Capps (chapter 4, this volume) illustrates how language is used by family members to frame an agoraphobic woman within a discourse of irrationality, thus collectively constructing and maintaining her identity as that of an unstable person. Much of this recent work on language as invention of self focuses on language used in live communicative contexts: Identity is performed, and jointly constructed, through choices about language forms. But narratives—oral or written, with the distinction between oral and written language practices more dubious than generally believed (see Shuman 1986)—provide another space in which identities can be constructed. Pam Gilbert argues that narratives "have a functional role in our culture: we live a good deal of our lives on the power of various stories, and it is through stories that we position ourselves in relation to others, and are ourselves positioned by the stories of our culture" (1992:186). Likewise, Capps (this volume) notes that in narratives we can see theories about emotions, self, and others, and about how past lived experiences are organized in "present, future, and imagined realms." Gendered literacy In a move away from strictly cognitive frameworks, some literacy researchers have begun to explore constructions of gender through both reading (Christian-Smith 1993; Davies 1989, 1993; Walkerdine 1990) and writing (Gilbert 1993; Kamler 1994; Phinney 1994). Judith Solsken (1993), for example, traces the experiences of four white middle-class children in relation to literacy over a period of three years, at home and school, highlighting children's unfolding processes of identity construction through the choices they make about reading and writing. Davies (1993) illustrates problems children face when they attempt to subvert traditional gendered narratives. As with much of the work on gender in general, research on gendered aspects of literacy has typically been conducted using white middle-class subjects or assuming a white middle-class norm. Researchers often call for attention to intersections of gender, class, race, and ethnicity but rarely explore them in practice. In this chapter, I examine gendered and classed aspects of the identities that Latina and Latino workingclass youth invent for themselves in their stories. I consider how the children position themselves through their characters within social discourses and how the discourses that are available to them are shaped by the social contexts in which they work.

66

IDENTITY AS INVENTION

Participants and setting The site of my study, Carol Lyons's mixed-age classroom, was located in a Latino working-class community just outside a large metropolitan area.1 Most of the children were immigrants or the children of immigrants from various regions of Mexico; a smaller number were recent immigrants from Central or South America. All of the students were fluent speakers of Spanish. Spanish predominated in most classroom talk, and all but a few of the books that the students used were written in Spanish. Carol was not a native Spanish speaker, but she was fluent in the language and used it in both her professional and personal life on a daily basis. Carol had taught in the school district for 10 years and was widely regarded as an excellent teacher within a child-centered, holistic tradition of instruction. Approximately 10 students in each of Grades 1, 2, and 3 were in the classroom in the first year of my study. The mixture of ages partially shaped the topics and content of students' books and the power dynamics that were established in authorship teams; where possible, I attend to these in the discussion that follows. All the second- and third-graders remained with Carol for the second year of the study, when the classroom was redesignated as a third- and fourth-grade combination; only the first-graders dropped out, to be replaced by eight additional third- and fourth-graders. In all, twenty-three boys and sixteen girls were in the room for some period of time during the 2 years, although there were never more than thirty-five total at any one time. Methods Thi s study, part of a larger ethnographic project, centered around understanding how the children invented identities in their written compositions. I engaged in participant observation several days each week for 13 months (across two overlapping school years) during the period devoted to the writing workshop and conducted eight follow-up visits over the subsequent 4 months. I compiled notes on 149 stories that were written during the first school year and 152 books from the second year and used these for analyses of topics, content, form, and structure. The notes included summaries of the story plots and verbatim excerpts from each text, as well as salient aspects of illustrations, design, and authorship. In addition, I talked informally with the students during every observation period and with Carol as demands on her time allowed. Carol and 1 met for two semiformal audiotaped interviews in her classroom and for extended conversations at her home and by telephone. Carol also read and commented on drafts of this work.

Inventing stories—Processes and practices Students wrote drafts for their stories during a reading/writing block in which they could choose from a variety of literacy activities. Writing and publishing books was the most popular activity, and a very social one, as the children actively sought out writing partners, sat together, and talked freely as they wrote. Like the kindergarteners in Margaret Phinney's (1994) study and the preschool children who were observed

Good Guys and "Bad" Girls

67

drawing pictures by Amy Kyratzis (1994), these elementary-school children used the writing workshop as a space for the negotiation of friendship groups and as an index of classroom popularity. Five or six of the oldest boys and one of the oldest girls participated in a large number of authorship teams. The three youngest boys almost always worked alone and the youngest girls worked alone or with each other as partners. There were very few mixed-sex authorship teams. In the first year, only three of the eighty-seven multiply authored books were written by mixed-sex teams; in the second year, only nine out of seventy-five. I witnessed few overt forms of "borderwork" (Thorne 1993)—social processes such as teasing that occur when gendersegregating norms are violated—but implicit rules seemed to keep girls and boys from writing together. This, in turn, may have led to more overt gender stereotyping in the content and characters represented in the stories. After drafts had been prepared, students met with peers and the teacher to offer ideas for revising or developing the text. In these conferences, Carol frequently emphasized the fact that all suggestions were "just ideas" and that each student author could decide whether or not to incorporate them into the stories. She generally limited her comments in order to open up more space for peer feedback. Nevertheless, she challenged students when they wrote about themes that reinforced gender or other stereotypes. After the writing conferences, students finished their stories and gave them to Carol for final editing and typing. The authors then pasted the typed text into premade booklets, illustrated them, and bound them in cardboard binders. The finished books were displayed in the classroom library and became the most popular reading material in the room. Excerpts of typical books are shown in figure 3.1. Most students did not verbalize how they came up with ideas for their stories. When I asked, most told me that they did not know or that they "got it from their heads." In general, however, peers' books appeared to exert a strong intertextual influence on new narratives. As new topics were introduced, they were picked up by others and played out in related texts over time. Overall, the students seemed more focused on the social than the literary aspects of the stories, and they negotiated the friendships that appeared in the stories much as they negotiated their own friendships in class. In their conversations with me and in letters that they wrote to their friends as part of the classroom mail system (another device Carol used to encourage literacy), many students mentioned the activity of coauthoring books as an index of friendship. This situation lends insight into why gender-segregated worlds appeared in so many texts, as discussed in a subsequent section. It also gives greater weight to my claim that students projected important aspects of themselves and their own desires through these texts; in students' books, the line between their actual and fictional lives often blurred. Inventing characters—Talk and text The degree to which the authors' lives melded with those of their characters is revealed by looking at the language forms they used to talk about texts and to position

Figure 3.1. Sample pages from a student text

Good Guys and "Bad" Girls

69

characters in their stories. They would invite friends onto authorship teams by asking, "Do you want to be in my story?" When writing, students would usually introduce characters using the third person, but after a few lines, they would frequently slip into the use of the first person and begin to speak in the voice of the characters rather than for them. One team wrote, for example: "Once upon a time there were three boys named Carlos, Marcos, and Fernando. They wanted to enter a car race. And so we asked our parents for permission. They told us we couldn't go." I was present when Carol attempted to call the class's attention to this inconsistency in perspective, and I sensed that the group did not understand Carol's concern. They seemed more interested in establishing the connection between their characters and themselves rather than signaling authorial distance from their characters. This quick shift in point of view is perhaps the most obvious example of how the children truly invented—and reinvented—themselves through their stories. The children also used language forms in ways that revealed their deeply gendered identification with the characters they created. When girls and boys did write stories together, they were extremely careful to distinguish between the language forms used for female and male characters in ways not required in the Spanish language. For example, Katia, Jesus, and Marco wrote a draft for a book called "Dos vampiros y una vampira" (Two male vampires and a female vampire), and Ofelia transformed "Los tres cochinitos" (The three little pigs) to "Dos cochinitas y un cochino" (Two little female pigs and a male pig). The children could have written about three generic vampires or pigs, leaving the gender ambiguous (or subsuming the feminine within the masculine, as is the rule in Spanish). Their careful language distinction may indicate personal identification with the characters and concern that they themselves be tagged with the proper gender label. I have argued elsewhere (Orellana 1995) that because the Spanish language is gender-marked, gendered dimensions of literacy in Spanish-dominated classrooms may be more evident (though not necessarily more prevalent) than they are in other language settings. But when biases are more overt, they are more readily challenged. The girls' literary practices described here can be seen as efforts to subvert linguistic patriarchy by writing themselves into the language. The authors call attention to linguistic forms that promote masculinist readings of texts, and they push readers to recognize that the characters in their stories are not the "generic" male. These efforts are not unlike those made by professional feminist writers to avoid masculinist biases in their novels (Livia 1994; chapter 17, this volume). Social relations within textual worlds The authors' intense identification with their characters suggests that the worlds they sketch in their stories and the characters they personify are intimately connected with their own identities and social worlds—both the actual and the possible, both the real and the invented—or at least with the identities and worlds with which they are willing to be associated on paper. The tensions that appear in the stories may illuminate the children's struggles to seek space for themselves in the world, to try on potential identities and explore the implications of various choices.

70

IDENTITY AS INVENTION

Gender-separate worlds Perhaps the most salient aspect of the children's stories was the extent to which they represented gender segregation. Girls were often completely absent from the boys' stories. Only one story written by a boy had a female lead character, and in only seven books were girls present as supporting characters. Women were somewhat more visible, appearing in a total of thirty-six of the boys' books, but mostly as mothers (who merely accompanied fathers in most of the texts); men appeared in fifty-two books in a range of positions, less than half of the time as fathers. Although girls favored girls or women for the lead figures (usually fashioning them as minimally fictionalized versions of themselves), they occasionally gave boys or men a central position in their writing. In the first year of the study, girls wrote two stories that centered around a single man and one that featured two boys in the lead, as well as four books with both a girl and a boy or several girls and boys as the central characters. In the second year, they wrote nineteen stories that revolved centrally around men or with male leading characters and two additional stories about mixed-gender groups. The girls also portrayed men or boys in supporting roles in sixty-eight stories. The fact that gender-separate worlds appear in most of the children's stories could suggest that the children value such segregation. However, although I did see some areas of clear gender separation in the children's work and play environments, I saw many other spaces in which the children mixed freely, both in the classroom and on the playground. In a sense, the children (the boys more than the girls) endorsed and sustained a discourse of gender segregation even as they disrupted or only partially maintained such segregation in practice. Support for this claim can be found in the literal—pictorial—representation of female and male characters, in which the influence of wider social discourses of gender is also apparent (figures 3.2 and 3.3). When women were featured in illustrations, they often appeared wearing princess-like dresses, with large, heart-shaped bosoms; in more than half of the colored illustrations they were blond with blue eyes. Boys and men were rarely represented as blond in either boys' or girls' illustrations, but in a number of the books they were clearly featured with blue or green eyes. Yet all of the students in the classroom had dark hair and eyes. This suggests that even as the authors identify closely with their characters, they sometimes construct images of themselves that are bound up with dominant racial discourses. However, as Rusty Barrett (chapter 16, this volume) notes, "the appropriation of aspects of dominant culture need not necessarily indicate acceptance of its dominating force" and can instead represent a form of resistance. In playing with a range of images for themselves, the children challenge stereotyped notions of what Latinos "should" look or be like. At the same time, these images seemed informed by a narrow range of possible gender positions. Thematic tensions In major themes that run through their stories, the students in Carol's classroom played with their portrayal of their own gendered and classed identities. In many texts, the children describe characters who are gendered in very particular ways and who stand

71

Good Guys and "Bad" Girls

Figure 3.2. Representations of women and girls in texts

at the extremes of particular dichotomies—as rich or poor, smart or dumb, blond or dark-haired, and good or bad. The authors present these characters in different lights and make assorted judgments on their fates. In doing so, they pick up dominant discourses about gender and class that exist in the world around them and that exert influence on their developing notions of self. At the same time, they also contest, disrupt, and refigure those differences in the worlds they create on paper. Love and romance Love, romance, and jealousy were central themes in many of the girls' stories and were a peripheral aspect of others. During the first school year, when the students

72

IDENTITY AS INVENTION

Figure 3.3. Representations of men and boys in texts

were in Grades 1 to 3, six stories by female authors revolved centrally around romance. For example, first-graders Janet and Mirna wrote about a female duck who ignored her mother's advice to find a rich man, instead following her heart, which led her to marry a poor male duck and have five babies. Four second- and third-grade girls joined forces to write a story about butterflies who fell in love and got married. These girls did not adopt Carol's conference suggestion that they consider alternative endings to the story, but they did seem to see through the illusion of living "happily ever after" when they remarked to Carol that caring for babies was hard and often frustrating work. The complex role of teachers in conferencing on such stories bears exploration. In suggesting alternatives to the "and-they-lived-happily-ever-after" ending, Carol

Good Guys and "Bad" Girls

73

aimed to expand the range of possibilities that the girls saw for themselves as projected through their characters. She always reminded her students that they could choose to accept or reject the suggestions that were made, but she consciously pushed students to explore options they had not considered. As she told me in our conversations, she wanted her students to critically examine the world and not to "accept all this unfairness in life." Yet this feminist challenge may echo institutional efforts to control young Latinas' sexuality and reproductive choices (Houghton 1995). Given that Latinas are effectively written out of dominant cultural narratives of romance, the girls' positioning can also be seen as a form of subversion of traditional norms. By contrast with the prevalence of romance in the girls' stones, only one story that dealt centrally with romance was written by a boy. Mario, a first-grader, wrote about a personified blue (male) car who married a pink female car (un carro muchacha); together they had purple baby cars. Mario was the same boy who wrote the only male-authored story to feature a female protagonist. In that story he wrote about a female elephant whose parents didn't love her because she was ugly; when she was not accepted in a ballet class, the elephant then joined a circus, where she taught ballet to the other animals. Mario may have been able to write against the grain because he was one of the youngest boys in the classroom who worked largely independently and hence never adopted the "hegemonic" forms of masculinity (Connell 1987) established by the older boys. In the next school year, when the children were in Grades 3 and 4 (with the firstgraders moving to second grade in a different classroom), fewer stories followed a love-marriage-babies formula. Two third-graders, Wendy and Mirna, wrote a book about a rose that fell in love with a carnation from Mexico; the flowers then had two babies—a rose and a carnation. Olivia, another third-grader, wrote a story about two mermaid friends who at first did not have boyfriends ("because they didn't want them," as she explained), but then found two handsome boys, fell in love, got married, and had babies. But in general, the girls' treatment of romance was more complex in the second year. Carie and Elsa, both fourth-graders, wrote about impossible love when they retold the story of a man who fell in love with a ghost; Katia, a thirdgrader, later reworked this idea, introducing a woman who fell in love with a male ghost but then stopped loving him when she realized he was a ghost, leaving the man pining away for her (and wanting to kill her). Several other stories evinced the influence of Spanish television novelas, either explicitly or implicitly following the themes of jealousy and betrayal that are central to that genre.2 Two books by fourth-grader Rosa hint at the tensions some girls may feel between operating within a discourse of love and romance and an alternative "good girl/good student" discourse. Love and romance may represent a sanctioned cultural pathway for girls, but they receive conflicting messages about when and how they are to embark on it, such that in certain times and places for girls to operate within a discourse of love and romance is to position themselves as "bad." This may be especially true for Latinas, who are often stereotyped in good/bad binaries, and for girls within the context of schooling, where being "good" generally requires a denial of one's sexuality. In the initial draft of her story, Rosa wrote about a girl who watched too many soap operas, talked about them incessantly in school, kissed a boy while under their influence, and performed miserably as a student until her teacher talked

74

IDENTITY AS INVENTION

to her mother. At that point the girl stopped watching novelas and became a model student. In a later version of the story, however, Rosa describes a girl who displays her resistance both to school and to being a "good girl"; she resists by watching television novelas, which deal centrally with love and romance and which position her within a highly gendered discourse. This leads to her failure to succeed in school and in life. The tensions here resonate with those described by Jean Anyon (1983) in her analysis of working-class girls' resistance to school and the ways that such resistance locks girls into traditional gendered and classed roles in life. Rosa displayed this same tension in the corpus of her writing. She wrote detailed accounts of love triangles, jealousy, and passion in her two versions of the story, but she was also the first in the class to write a story railing against the actions of a serial rapist who was at large in the town at the time. She additionally wrote a series of stories and journal entries about strong women in her life: her mother (whom she referred to as la mama valiente or 'the brave mother'), her teacher, the principal, and myself. In none of these writings did she describe women in relation to men; however, beauty and goodness were the chief traits highlighted in each. In these works, Rosa seems to oscillate between playing out romantic fantasies and protesting against the harsh realities of women's experiences. Good Guys, Bad Guys . . . As may be evident in Rosa's stories about love and romance, a second major tension that ran through the children's writing centered around what it meant to be good or bad. This tension became gendered because goodness and badness were inscribed differently for girls and for boys, and the children seemed to use the stories to play out different ways of being gendered and to consider the implications that followed from each. Most of the lead characters in the girls' stories from the first phase appear to be "good girls"; they are obedient, submissive, passive, thoughtful, cheerful, kind, and peaceful. They either stay at home or go to a few places with friends, live happily with their families, and contemplate beauty in the surrounding world. They live, immobile, in homes that have few problems and that seem detached from the world around them, existing in a sort of idyllic social vacuum. When conflicts do arise, as in Denora and Noemf's story about a beautiful house that gets broken in a fight, people come back together to patch things up. Problems are resolved and the protagonists can live happily ever after. Such stories contrast with those of the boys. In the boys' first stories, there are a few images of "good boys," who appear as compliant, hard workers and obedient children, but more often the boys assume much more powerfully good roles, when they become "Good Guys" who fight the forces of evil. Good Guys represent a rather different construction than good boys, in that they arc bold, powerful, authoritative figures, while good boys suggests an image of quiet, respectful children who do as they are told. In these stories, however, the Good Guys get to get in on some really "bad" action; the theme seemed to allow the boys to be both good and very powerful through character traits that were not as accessible to girls. Many of these books also 'allowed the boys some vicarious participation in the deeds of the Bad Guys without

Good Guys and "Bad" Girls

75

having to become bad themselves; the authors used their own names for the heroes in the stories and set up the Bad Guys as "Others" but often provided elaborate details of the Bad Guys' dark deeds, and they sometimes wrote avidly about war even as they made their characters resolve their conflicts by waging peace. This may have allowed them to please two different audiences: their peers, who were as intrigued by war and "badness" as they were, and their teacher, who overtly promoted peace. It is worth noting that the characters who were transformed into superheroes were often explicitly constructed as poor boys, and when they became superheroes they specifically used their powers to help the poor. Through these superhero characters, the boys may have found a way to assume power in a world that typically denies it to working-class children and to rectify some of the inequalities they sensed in the world around them. Boys from lower-class communities may feel their lack of formal social power more keenly than girls because public power is generally framed as masculine. In other stories, the boys did not transform themselves into superbeings in order to play out powerful roles; they found other means to achieve similar ends. For example, fourth-grader Fernando wrote about a poor boy who overcomes great odds by studying hard, earning a scholarship to Yale, and joining the army. There he assumes a powerful position but uses his power to stop the war. Although none of the girls ventured into the realm of superhero stories, several female authors did experiment with more active heroine-like roles in books produced during the second year of the study. For example, Carie, Vicky, and Elsa joined together to write "Las tres detectivas" (The three female detectives), a story in which the authors are the only detectives who are able to capture an international thief. In "Las angelitas" (The little angels), Carie takes this interest in powerful roles to new heights; she assumes an almost omnipotent role as a poor girl who gets killed but then unmasks the devil, defeats him in a fight, and liberates God himself. In the end, Carie celebrates with her friends because God is happy with them. Rosa also experimented with at least one powerful role for herself, along with the strong characters she gave other women in the biographical sketches she wrote. In "Las tres mosqueteras" (The three musketeers), Rosa, Elsa, and a third female character (named for the teacher) ride on three horses (named for three of the most "popular" boys in the classroom) and go out to save the king's son at his request. The story ends before we find out if their mission is successful, however, and the authors never produced the promised continuation. 3 Of all the girls in the room, it is not surprising that Rosa and Carie were the ones who experimented most with powerful portraits of women and girls in their writing. They were two of the most outspoken students in the room and held their own in public spaces that were largely dominated by boys, such as classroom meetings in which students presented their grievances (Orellana 1996). Other students in the classroom also often followed Rosa's and Carle's leads. Nevertheless, these girls experienced tension about their positions in the classroom and in their texts. Rosa and her coauthors, after all, left open the question of whether or not they save the prince at the end of their story. If they opted to save him, they would frame themselves as powerful but would play into a traditional gendered story line (albeit in a reversal of roles). Opting not to save him could be read either as a feminist decision or as evidence of "feminine" weakness. The decision not to choose may be the most radical

76

IDENTITY AS INVENTION

one these girls could make under circumstances set up for them by societal expectations. Similarly, although Carie defeats the devil in "Las angelitas," she frames this act as done to make God happy (which could be construed either as a very powerful act or as father-pleasing), and she did not protest when her male coauthors assigned her the role of cheerleader in a story about a football team. Through their stories, the children in this classroom seem to acknowledge that there are very different options for being good open to girls and boys, with different implications in terms of power, prestige, and influence on the world. The fact that a few of the girls did write adventure stories in which they present themselves as heroines suggests that there is some room for girls to cross this gender-marked border and some motivation for them to do so. Conversely, there were no instances of boys taking up a theme that had been established by the girls. . . . and the lure of being bad This interest in "goodness" in the early girls' stories was not matched in the boys' stories. The boys seemed as fascinated by "badness" as they were by "Good Guyness," and they wrote a number of books whose main characters were constructed as bad boys. Some of their earliest narratives on this theme took on moralistic tones, ending with the boys rectifying their ways or meeting their just fate (as in Fernando's account of a man who takes drugs, becomes an addict and a thief, and ends up in jail). Other stories, however, clearly suggest the boys' temptation to be "bad" and to display themselves in socially taboo ways. Two books were written about the super pedoros (super farters), in which boys who fart have a series of adventures but ultimately die in a fire started from their own fumes; and another was written about the super traviesos (super menaces)—twin boys who engage in a series of mischief. In the second phase of the study, the boys experimented with ever bolder characters. In Fernie and Robby's "Los alrevezados" (The mixed-up ones), the bad boys come from another planet, steal things on Earth, laugh at others, live on the street, and eventually die from their own stench. In Carlos's "El super travieso" (The super menace), a boy who fights both in the United States and in Mexico gets kicked around the world (literally) as a consequence of his bad behavior. In Robby, Julio, and David's story about three boys who fight in school and get suspended, only one of the boys (the "little angel") is sorry for his deeds; the other boys seem confident in the power of their badness. Robby and Julio also wrote "Los super gordos" (The super fat ones) and "Los super tontos" (The super dumb ones), stories in which the authors seem to delight in having the characters engage in socially taboo activities such as farting, saying bad words, and defecating on the seats of an airplane. Later, Carlos, Robby, Jorge, and Fernie joined forces to produce a story that plays with taboo subjects in an especially provocative way: In "Los cuatro draculas" (The four Draculas), the spirit of the mother of four vampire boys appears on the scene, farting and calling to her sons as the character La Llorona docs in the classical Latin American folktale. The boys order their mother to make underwear soup (sopa de chonis) for them to eat. It is worth noting that in this story, the boys selected a female student in the class, Sarai, to play the less-than-flattering part of the mother who farts (and who abandons her children by dying). This situation raises the question, also raised by Timo-

Good Guys and "Bad" Girls

77

thy Lensmire (1994), of how writing workshop environments may allow authors to position their classmates in oppressive ways. The authors do not name one of themselves as the father in the story, perhaps because they are not willing to be associated in a textual marriage and subsequent pregnancies with Sarai. Although they do not name themselves as the children, the fact that there are four vampires and four authors is suggestive of their identification with that role. By the third month of the second school year, however, it was not only the boys who were experimenting with having their main characters do really "bad" things. Three girls (Denora, Noemi, and Elsa) wrote about boys who engage in such socially unacceptable behavior as throwing the baby Jesus on the floor in a Christmas pageant and putting screws on each other's chairs. Still, in these texts the authors distance themselves from the characters and present clear consequences to the boys for their badness. In the first story, the "bad" boy does not receive any gifts from Santa Claus; during the same time frame, Denora wrote another book in which a "good" girl gets many presents for Christmas. Seven other girls, however, took the theme of "badness" and made it their own. In different groupings over a period of several months, they produced at least five stories in which they themselves were featured being "bad," in several cases engaging in taboo behavior that rivaled that portrayed by the boys. For example, Vicky and Sarai wrote about girls who rob a store to get stickers after they fight over the stickers they have bought. Carie, Elsa, Sarai, and Mirna wrote about three girls who steal money, rob a store, and play with meat in a meat shop. Sarai, Laura, and Mirna wrote about girls who fight with each other and break each other's things. Sarai, Mirna, Wendy, Elsa, Carie, and Cindy joined forces to write about two girls who try to strangle each other, make jokes about bodily functions, and destroy things in their house. Sarai was an author of each of these stories and seemed to be a principal force behind their more provocative aspects. When she was engaged with Laura and Mirna in writing "Las ninas traviesas," she seemed to delight in narrating to me: "Then she (the sister) is gonna kick her butt, and then she (the other sister) is gonna kick her butt. Like that." In general, Sarai presented herself as outgoing, confident, bold, and friendly. The other girls were quiet, soft-spoken students who projected an image of prototypically good girls: They rarely spoke publicly in class, they did their work diligently, and they never broke the classroom rules. Yet Mirna and Wendy by themselves wrote a fifth story about mischievous girls, in which they portrayed a struggle over what it means to be good or bad: They appeared as girls who destroy their mother's things until she sends them away to obedience school. When they return they are "good"—they clean the house (while their mother cooks and then watches television). Then, however, the girls watch a show on television about two mischievous girls and once again become "bad" and destroy the house; they remain "bad"— and powerful in their destructive forces—at the end of the book. Intersections of poverty, wealth, power, and gender As indicated in my previous discussion of story themes, references to poverty and wealth were prominent in the children's stories. The boys, in particular, wrote about winning money in car races and boxing matches; they wrote about poor boys who

78

IDENTITY AS I N V E N T I O N

are transformed into superhcroes who become rich; and they made frequent references to los pobres (the poor). By using this term, or by presenting themselves as poor boys who escape poverty, the boys distance themselves from poverty even as they identify with it; treat it as a condition that can be overcome by those who are strong, smart, or talented; and help to construct an image of those who remain in poverty as nameless, faceless "Others." Just as in their treatment of goodness and badness, the children represented poverty and wealth as dichotomous, with transformations between the two possible but with little room for middle positions. The constructions of poverty, wealth, and power are in many ways bound up with constructions of goodness and badness, which are in turn interwoven with constructions of gender, so much so that it is difficult to separate the threads of these themes in my discussion. Yet in the interrelationships between the dimensions of poverty/wealth, goodness/badness, and gender displayed in the children's writing, the most important insights into their understanding of social identities may be gained. For example, in only one story is a bad child specifically designated as rich, yet in more than a few stories good persons are, and other poor boys become both good and rich when they assume super power. Many of the bad children in the stories project an image of poverty, as indicated by homclessness. This portrayal suggests that the children may see poverty as a sort of punishment for being bad; the poor whom they see around them (and of which group, statistically at least, they are part) somehow deserve their fate. This interpretation is supported by the fact that in the single story in which a rich boy is "bad"—Fernando, David, and Walter's "El saborrion, el enojon, y el presumido" (The glutton, the mad one, and the presumptuous one)—the rich boy dies and loses all his money to the other boys in the story, who are poor, good, and not presumptuous, and so are apparently rewarded for their goodness. This story line has important implications for the boys' self-images—they seem to present themselves as good and powerful only when they deny or escape their current social-class positions. The girls did not seem as concerned with issues of poverty and wealth as did the boys; they generally dealt only peripherally with the theme, with a few exceptions. Denora, Noemi, and Elsa wrote a story about a poor, homeless girl who entered a singing contest and won three million dollars. The girl didn't know what to do with so much money; she bought a house, grew up, got married, and "vivio muy feliz" (lived very happily). Later, the same girls wrote a fictionalized biography of the popular singer Gloria Trevil, who really did get rich by singing. In one other story, Denora and Noemi wrote about eight orphaned rich girls who are poisoned on Halloween, but the girls' wealth is only an incidental part of the story line.

Summary and conclusions The students in this classroom can be seen inventing social identities through their narrative writing, where they write the word, the world, and their very selves. They write themselves into a range of possible existences and create multiple identities for themselves, even as their freedom to create is constrained by the realms of what

Good Guys and "Bad" Girls

79

they view as possible fictional lives. Some of the identities that students invent for themselves in their stories correspond closely with some of their actual behaviors in the world, and some do not. Some aspects of their identities are directly represented in print, and others are transformed, modified, denied, or erased. And some of the personas that individuals assume in different stories contradict each other: They may be Good Guys in one story and bad boys in the next; they may be girls whose main dream is to marry a handsome man in one story and musketeers going on heroic missions in the next. Yet such contradictions are inherent in the multiplicity of beings that all of us are or at times dare to be. The "free-writing" process that was cultivated in this classroom allowed room for the invention of multiple identities in ways that are not open to students in more traditional classrooms, where writing topics are assigned or strict regulations are placed on what students can write about. Writing workshops such as this one may provide one of the only legitimate academic arenas in which students beyond the preschool age are able to play: to experiment with a range of characters and stories and try on a number of possible social identities. The stories became a place where the children could be rich, or powerful, or really good, or really bad. They could push the limits of social regulation, as when they wrote about taboo topics without any censure by their teacher. In doing so, they could at times cross borders such as those defined by gender. For the girls, the opportunity to be bad may be especially significant, given the strong pressures that are typically exerted on girls to be good and the ways in which their construction as good girls can repress other aspects of their identities, as Valerie Walkerdine (1990) suggests. Yet in general the students did not voluntarily cross gendered borders in their writing, and when they did, their crossings were not equilateral. Both girls and boys created largely gender-separate worlds in their texts and generally inscribed their characters within supposedly gender-appropriate roles, but girls were more likely to move into territory that was defined by boys than vice versa. Several girls joined boys' groupings and took up topics that had been established by the boys, whereas only one boy, Mario, explored more overtly "feminine" topics such as love, romance, and ballet lessons. Thus, as the work of Gilbert (1992) and others would suggest, the "freedom" of this free-writing environment may be illusory; the fantasy worlds that the children created on paper were largely influenced by the social worlds that they saw around them and by the social discourses that regulate action in that world and define the range of possible identities that the children could safely try on. The forms of border-crossing seen in the stories also deserve consideration. I found myself asking what it means for girls to break out of the strictures of the "good girl" mold by choosing to be bad. From a feminist perspective, is it necessarily "bad" to be "good"? Similarly, what does it mean when boys from low-income communities display in their writing their internalization of a work-ethic ideology which suggests that anyone who works hard enough or is smart enough can get rich? There are dangers in romanticizing resistance or in seeing all forms of subversion as liberatory. As Paul Willis (1977) and Jean Anyon (1983) demonstrate, working-class adolescents' resistance to the regulatory forces of school may help lock them into deeply gendered working-class lives, and resistance may serve to reproduce and reinforce the given social order. At the same time, as outsiders to these intimate processes of

80

IDENTITY AS INVENTION

identity construction, we should be equally careful not to see children's expression of gendered and classed identities—even seemingly stereotypical ones—as evidence of their submission to the given order or as a sign that resistance is "really" reproduction. It may be a sign of both or of neither. For researchers and theoreticians, the issue becomes one of how to interpret the identities that children invent for themselves within existing discursive—and structural—power relations. For teachers and activists, the question is how to respond to children's narratives in order to open up the widest range of possibilities to all students, how to encourage acts of subversion without reinforcing mere reactions and without imposing new limitations on the identities children are allowed to invent. The struggle is one of seeing, and helping our students to see, the complex mechanisms that maintain class and gender divisions in society while still exploring how to pursue our own dreams, invent our own futures, and write our own lives into existence in the world. I would suggest that the best answer lies in seeing that no one is locked into a single position on either end of a system of binary opposites. We can all be both good and bad, both strong and weak, both students and lovers. We can speak from different positions simultaneously or choose where and when we want to fit into a particular set of discursive—and human—relations. We can invent ourselves again and again, in different ways in various places and times, and in doing this we can help challenge the forces that limit the realms of our real and invented worlds. NOTES

Thanks to the teacher and students for sharing their classroom, ideas, and writing with me; thanks also to the editors of this volume, Laurie MacGillivray, and Barrie Thorne for thoughtful feedback on earlier drafts. 1. Pseudonyms have been used for the teacher, the students, and the school. In this chapter I refer to the teacher by her first name because that was her preferred form of address; however, most students called her "Miss Lyons" or "maestra" (teacher). 2. Spanish television novelas, while reminiscent of English-language soap operas, represent a distinct genre with their own historical and cultural traditions and should not be equated with their U.S. counterparts. A full discussion of this genre is beyond the scope of this chapter, however, as is an adequate discussion of the complexities of romantic genres and their reception. See Mary Bucholtz (chapter 18, this volume), Linda Christian-Smith (1990, 1993), Janice Radway (1984), Laurel Sutton (chapter 8, this volume), and Carol Thurston (1987) for discussions of "feminine" genres and their readings. 3. The lack of closure in this and other stories written by the students in this room merits investigation. On the one hand, it may represent the easiest way out of the uncertainty of how to end stories. This may be particularly important when any such decision would have moralistic overtones or would require the authors to make a statement about what they believe is proper, right, and good. By leaving the story inconclusive, the onus of the moral decision is removed from the authors and shared with the audience. At the same time, given the authors' close identification with the characters in their stories, the lack of closure may be motivated by an attitude such as that described by Kathleen Wood (1994:777) in her analysis of lesbian coming-out stories: Like the narrators in Wood's study, the children may not have "one fixed and rigid understanding of their lives"—or the lives of their characters—and so

Good Guys and "Bad" Girls

81

they leave the audience with a sense that there is more but that they themselves may not be privy to that part of their characters' lives. REFERENCES

Anyon, Jean (1983). Intersections of gender and class: Accommodation and resistance by working class and affluent females to contradictory sex-role ideologies. In Stephen Walker & Len Barton (eds.), Gender, class, and education. New York: Palmer Press, 19—38. Atwell, Nancie (1987). In the middle: Writing, reading, and learning with adolescents. Upper Montclair, NH: Boynton/Cook. Brooke, Robert E. (1991). Writing and sense of self: Identity negotiation in writing workshops. Urbana, IL: National Council of Teachers of English. Calkins, Lucy McCormick (1986). The art of leaching writing. Portsmouth, NH: Heinemann. Christian-Smith, Linda K. (1990). Becoming a woman through romance. New York: Routledge. (ed.) (1993). Texts of desire: Essays on fiction, femininity, and schooling. Washington, DC: Palmer Press. Connell, Robert W. (1987). Gender and power. Stanford, CA: Stanford University Press. Cook-Gumperz, Jenny (1995). Reproducing the discourse of mothering: How gendered talk makes gendered lives. In Kira Hall & Mary Bucholt/, (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 401-419. Davies, Bronwyn (1989). Frogs and snails and feminist tales: Preschool children and gender. Boston: Allen & Unwin. (1993). Shards of glass. Cresskill, NJ: Hampton Press. Dyson, Anne Haas (1989). Multiple worlds of child writers'. Friends learning to write. New York: Teachers College Press. (1993). Social worlds of children learning to write in an urban primary school. New York: Teachers College Press. Gal, Susan (1995). Language, gender, and power: An anthropological view. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 169-182. Graves, Donald (1983). Writing: Teachers and children at work. Exeter, NH: Heinemann. Gilbert, Pam (1992). Narrative as gendered social practice: In search of different story lines for language research. Linguistics and Education 5:211-218. (1993). (Sub)versions: Using sexist language practices to explore critical literacy. Australian Journal of Language and Literacy 16(4):323-332. Houghton, Cathryn (1995). Managing the body of labor: The treatment of reproduction and sexuality in a therapeutic institution. In Kira Hall & Mary Bucholtz (eds.). Gender articulated: Language and the socially constructed self. New York: Routledge, 121—142. Kamler, Barbara (1994). Gender and genre in early writing. Linguistics and Education 6:153182. Kyratzis, Amy (1994). Tactical uses of narratives in nursery school same-sex groups. In Mary Bucholtz, A. C. Liang, Laurel A. Sutton, & Caitlin Hines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and language Conference. Berkeley, CA: Berkeley Women and Language Group, 389—398. Lensmire, Timothy (1994). When children write: Critical revisions of the writing workshop. New York: Teachers College Press. Li via, Anna (1994). The riddle of the Sphinx: Creating genderless characters in French. In Mary Bucholtz, A. C. Liang, Laurel A. Sutton, and Caitlin Hines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 421-433.

82

IDENTITY AS INVENTION

MacGillivray, Laurie, & Ana M. Martinez (1998). Princesses who commit suicide: Primary children writing within and against stereotypes. Journal of Literacy Research 30(10):53-84. Orellana, Marjorie Faulstich (1995). Literacy as a gendered social practice: Tasks, texts, talk, and take-up. Reading Research Quarterly 30(4):674-708. (1996). Negotiating power through language in classroom meetings. Linguistics and Education 8:335-365. Phinney, Margaret Yatesvitch (1994). Gender, status, writing, and the resolution of kindergarten girls' social tensions. Linguistics and Education 6:31 1-330. Radway, Jane (1984). Reading the romance. Chapel Hill: University of North Carolina Press. Shuman, Amy (1986). Storytelling rights: The uses of oral and written texts by urban adolescents. New York: Cambridge University Press. Solsken, Judith W. (1993). Literacy, gender, and work in families and in school. Norwood, NJ: Ablex. Thorne, Barrie (1993). Gender play: Girls and boys in school. New Brunswick, NJ: Rutgers University Press. Thurston, Carol (1987). The romance revolution: Erotic novels for women and the quest for a new sexual identity. Urbana: University of Illinois Press. Walkerdine, Valerie (1990). Schoolgirl fictions. New York: Verso. Willis, Paul E. (1977). Learning to labour: How working class kids gel working class jobs. Farnborough, England: Saxon House. Wood, Kathleen (1994). Life stories as artifacts of a culture: Lesbian coming-out stories. In Mary Bucholt/,, A. C. Liang, Laurel A. Sutton, & Caitlin Hines (cds.), Cultural performances: Proceedings of the Third Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 777-786.

4

LISA CAPPS

Constructing the Irrational Woman Narrative Interaction and Agoraphobic Identity

his chapter illuminates the construction of irrationality through analysis of family interaction involving a woman ("Meg") who identifies herself and is identified by clinicians and those around her as agoraphobic. In Meg's words:1 I was diagnosed [with agoraphobia] two years ago . . . and now I identify myself first and foremost as agoraphobic. Fear is constantly in my mind. . . . I feel "less than" other people, kind of crazy somehow. And I know my kids and my husband, who's the world's nicest, most normal guy, must think of me as irrational. I'm somebody who's afraid to do most EVERYthing that normal people do without thinking twice. Just thinking about normal, everyday things . . . can send me into a panic.

Irrationality is widely considered to be a central characteristic of agoraphobia, which is most commonly diagnosed in women. Although genetics clearly contributes to the development of the disorder (e.g., Weissman 1993), environmental factors also play a significant role. Various theories have been proposed, the majority of which conceptualize the problem as existing within the mind of the individual. Further, researchers—even those examining hypotheses concerning the interpersonal dimensions of agoraphobia—typically employ measures designed to identify traits that constitute the "agoraphobic personality." Indeed, Meg's self-portrait is emblematic of this point of view as she identifies herself as "agoraphobic," "less than," "irrational," and "crazy" and differentiates herself from "normal" people. This study offers an alternative view of psychopathology in general and agoraphobia in particular, a perspective grounded in the notion that psychological disorders are interactional achievements that cannot be divorced from a particular sociohistorical 83

84

IDENTITY AS INVENTION

environment (Bateson 1972; Sass 1992; Szasz 1974). From this point of view identities are not static, individual entities. Rather they are dynamic, collaborative constructions that emerge moment by moment over the course of social interaction. Thus understanding agoraphobia requires examining interactions between people as they unfold over time and locating these interactions within a particular sociocultural context. As described in the American Psychiatric Association's (1995) Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), individuals with agoraphobia suffer irrational fear of being in a place or situation where it may be difficult to escape or to obtain help in the event of a panic attack or other potentially incapacitating or extremely embarrassing symptoms. A principal feature of agoraphobia is avoidance in response to this fear. Agoraphobic persons often describe feeling trapped by an ever-present threat of panic and their belief that they cannot risk leaving safe havens such as their homes. Although the term agoraphobia means 'fear of open spaces', the disorder is perhaps more appropriately described as a fear of being anyplace where one might feel alone and vulnerable to fear and panic. Initial reports suggested that agoraphobia is extremely rare, a conclusion that is likely due in part to agoraphobic persons' reluctance to travel to therapy sites. In fact, the disorder is more prevalent than previously suspected. The most recent epidemiological study of psychiatric disorders in the United States conducted in-home interviews of 15,490 individuals from five metropolitan areas, 4 percent of whom reported having suffered from agoraphobia during the previous year (Eaton, Dryman, & Weissman 1991).2 Although the disorder is more common in urban than rural areas, its prevalence appears to be similar throughout industrialized countries, across ethnic groups, and across age groups between 18 and 64 years. The epidemiological study found that agoraphobia is more than twice as prevalent among women as among men. Though compelling, this ratio is considerably lower than those found in earlier investigations of agoraphobia, which reported that up to 95 percent of sufferers are women (Marks & Herst 1970). In the clinical literature, agoraphobia is consistently referred to as a "woman's syndrome" (e.g., Foa, Steketee, & Young 1984). Discussion of the association between agoraphobia and women has focused not only on prevalence rates but also on the similarity between symptoms of agoraphobia and stereotypical female gender roles. Conventional gender identities, after all, are not inborn; they are socialized. Moreover, language is central to the socialization process, in that expert members of society use language to socialize children and novices into particular identities, roles, and worldviews, the expression of which involves using language in certain ways (Ochs 1988; Ochs & Schieffelin 1984; Schieffelin 1990). A considerable body of research, for example, suggests that women are socialized into indecisive, indirect, and deferential communicative styles relative to the more direct, authoritative styles of men (Lakoff 1990). From a feminist perspective, such patterns of communication both reflect and constitute women's status as powerless. Enactment of conventional female roles, whether by women or men, is also often considered pathological. Indeed, neurotic symptoms have been conceptualized as an extreme form of indirect communication (Breuer & Freud 1957; Szasz 1974). Building on the premise that the powerless have more incentive than the powerful to use indirect strategies, feminist and critical theorists have concluded that the association between conventional-

Constructing the Irrational Woman

85

ized feminine actions and stances and insanity or irrationality is a means of social control (e.g., Chesler 1972; Foucault 1965; Lakoff & Coyne 1993; Wenegrat 1995). With respect to agoraphobia, feminist theorists propose that traditional ideologies of femininity are "phobogenic," such that women are socialized into dependent, unassertive, and accommodating behaviors that are characteristic of the disorder (Fodor 1974; Wolfe 1984). Surprisingly, however, studies designed to empirically evaluate these socially situated conceptualizations of agoraphobia have relied on static, individualized measures that look through language, not at language in social interaction (Capps & Ochs 1995a, b). For example, to test the sex-role theory of agoraphobia, researchers have administered personality inventories, including masculinityfemininity scales, to individual sufferers. Results generally suggest that both women and men with agoraphobia rate lower on indices of masculinity (such as assertiveness and independence) than does a normative sample (Chambless & Mason 1986; Kleiner & Marshall 1987). Although consistent with the sex-role theory, this methodological approach locates the problem within individuals' personalities and does little to illuminate relevant sociocultural processes. Similarly, it has been proposed that people with agoraphobia wish to escape a close relationship but fear independence (e.g., Goldstein & Chambless 1978; Hafner 1982). Because the disorder confines sufferers to their homes, agoraphobia negates the first option and thus eliminates the conflictual situation. Many have dismissed this theory based on results of standardized questionnaires administered to persons with agoraphobia and their spouses which suggest that such couples experience no more marital distress than normal controls and that spouses of persons with agoraphobia show normative levels of psychopathology. These findings have led some researchers to claim that "agoraphobia derives from the individual patient's psychological make-up rather than the marriage or the husband, since the problem lies not in the marriage, nor in the husband, but within the patient herself" (Symonds 1973, as quoted in Arrindell & Emmelkamp 1986:600). Rendering agoraphobia an intrapsychic disorder, this statement makes explicit an ideology that is implicit both in the DSM-IV, which diagnoses disturbances in individuals, and in research that relies on individualized measures.3 In contrast to such assumptions, the following analysis is grounded in the notion that understanding psychological dispositions requires closely examining social interactions between people rather than the decontextualized responses of isolated individuals. The basic tenets of the present study are that people verbally attempt to establish their emotions, actions, and identities moment by moment during social interaction and that narrative provides a powerful resource for doing so (e.g., Bruner 1990; Heath 1983; Miller, Potts, Fung, Hoogstra, & Mintz 1990; Ochs 1993; Schieffelin 1990). In narrating, interlocutors attempt to construct themselves from a particular point of view, both as protagonists acting and feeling in the past and as narrators acting and feeling in the present. Narrators describe a setting in which a protagonist encounters a problematic event of some kind and relate ensuing psychological and behavioral responses and consequences. By forging causal connections between emotions and events, narratives build theories about experiences (Bruner 1986; Feldman 1989; Ochs, Taylor, Rudolph, & Smith 1992; White 1980). Further, a single telling can build conflicting theories about an experience or highlight conflicting dimensions of a protagonist's or narrator's identity (Capps & Ochs 1995a, b). Narrative interactions

86

IDENTITY AS INVENTION

create opportunities for negotiating identities and worldviews, for resisting, challenging, and perpetuating the status quo. In this sense, narrative theories not only shape understandings of past events but also organize lived experience in present, future, and imagined realms (M. H. Goodwin 1991; Ochs 1993). Narratives told in the course of social interaction are jointly produced by copresent interlocutors; that is, emergent narratives are interactional achievements (C. Goodwin 1981; C. Goodwin & M. H. Goodwin 1992; Jacoby & Ochs 1995). For this reason, individuals' attempts to construct a particular evaluative perspective in narratives require ratification from others. Conarrators also construct the identities of participants by shaping their roles as protagonists acting in the past and narrators acting in the present. Copresent interlocutors influence the unfolding of storytelling sequences through their gaze patterns, body orientation, and verbal contributions to the storytelling. In these ways, cotellers simultaneously display varying degrees of alignment with the initial teller's evaluation of the narrated events. In so doing they validate or repudiate the teller's legitimacy as narrator. Storytelling interactions provide a medium for the construction of irrationality in that being "rational" involves having the authority to reframe experience in a way that is ratified by others, whereas being "irrational" involves failing to receive such ratification from legitimate others. The pain and isolation that accompanies mental illness is partly attributable to loss of the authority to reframe events and experiences in ways that will be ratified and loss of confidence in one's ability to regain this authority. This is particularly true of people suffering from agoraphobia, for whom familiar, seemingly innocuous territory away from home keys the threat of panic and helplessness. Meg articulates this predicament in the passage quoted at the beginning of this chapter: "Just thinking about normal, everyday things... can send me into a panic." The construction of rationality and authority through storytelling is tied to the distribution of narrative roles. Elinor Oehs and Carolyn Taylor's (1992a, b, 1995) analyses of storytelling during family dinners have shown that members assume particular narrative roles and that these roles discursively arrange protagonists and interlocutors in relationships of power. Specifically, mothers in their study tended to introduce narratives, the majority of which featured themselves and their children as protagonists, whereas fathers most often served as primary recipients of the stories. By directing children to tell fathers about an incident in their day or by initiating narratives about their own personal experience, mothers nominate their children and themselves as protagonists whose actions are accessible to other interlocutors, particularly fathers, to review and sanction. Similarly, those who feel themselves to be irrational are likely to seek validation from others for the way they frame or make sense of their experiences. One way of doing so is by initiating stories about their own experiences so that they might be ratified by "rational" others. The act of designating a particular cointerlocutor as preferred responder constructs this recipient as rational and places her or him in a position to ratify the teller's explanation of a particular set of events. The present study illuminates how family storytelling interactions construct the irrational woman by examining (1) the distribution of narrative roles (i.e., protagonist, initial teller, primary recipient) among members of a family in which the mother (Meg) has been diagnosed with agoraphobia; (2) other family members' contributions (e.g., gaze patterns, body orientation, and verbal contributions) to stories that Meg initiates about her own disturbing encounters and to stories in which Meg is not the narrative

Constructing the Irrational Woman

87

protagonist or principal teller; and (3) how narrative interactions might fuel the perpetuation of "irrational" panicky thoughts and feelings associated with past or hypothetical events, which in turn may perpetuate Meg's identity as agoraphobic.

Methodology Data collection Participants. The present study focuses on narrative interactions between members of a middle-class European American family (the "Logan" family), including a mother/wife ("Meg," age 36); father/husband ("William," age 38), daughter ("Beth," age 11) and son ("Sean," age 6). Meg was initially diagnosed with agoraphobia by a clinician in her community. Her symptoms increased over the subsequent 2 years, at which time Meg started a self-help group for persons with agoraphobia. As a graduate student in clinical psychology, I approached members of the group to recruit participation in a dissertation project on the psychological adjustment of children of agoraphobic parents (Capps 1996; Capps, Sigman, Sena, Henker, & Whalen 1996). When Meg expressed interest in the study, I evaluated the appropriateness of her participation by confirming her diagnosis of agoraphobia using the Anxiety Disorders Interview Schedule, or AD1S (DiNardo, O'Brien, Barlow, Waddell, & Blanchard 1983). After participating in the dissertation study, however, both Meg and her daughter Beth expressed concern that the methods used did not capture Meg's experience. Meg and the other members of her family then agreed to embark on the present discourse-analytic venture. Corpus. The database for this study was collected as part of an in-depth ethnographic and discourse study carried out on the construction and socialization of agoraphobia in a single family (Capps & Ochs 1995a, b). Data collection consisted of (1) 36 months of participant observation of family life, including video and audio recordings of Logan family dinner interactions and leisure activities; and (2) audio-recorded, loosely structured interviews with Meg, alone and together with Beth. Data analysis In this study narrative is defined as a socially organized conventional telling of temporally ordered events in one's life from a particular evaluative perspective (Goffman 1959; Labov & Waletzky 1968). Seventeen narratives were gleaned from videotaped observations of dinner interactions, which served as the basis for determining the distribution of narrative roles. Subsequent analyses of family interactions focus on three stories in particular, which were selected in order to investigate how the irrational woman is coconstructed through storytelling interactions that involve family members in different roles. The three stories closely examined here are also thematically related in that each features a pit bull terrier, a notoriously ferocious dog: the first pit bull story is initiated and principally told by Meg, and the narrative centers on her own anxious experience; the second is initiated and principally told by Meg's

88

IDENTITY AS INVENTION

husband William and provokes anxiety in Meg in her role as coteller; and the third story is initiated and principally told by Meg and centers on another female protagonist's distress.

Narrative roles Examination of the stories reveals that the roles family members assume in narrative activity shape the identities that emerge. Although narratives are collaboratively constructed by all copresent interlocutors (Duranti 1986; Goodwin & Duranti 1992), family members assume different roles in the telling. Storytelling interactions organize protagonists and tellers in relations of power, which shape the identities that emerge. The recurrent arrangement Ochs and Taylor (1992a) identified in their study of family dinner activity establishes the father's position as "'panopticon'—the all-seeing eye of power" (Ochs & Taylor 1992a:329), a term taken from Michel Foucault (1979). As applied to narrative structure, the concept of the panopticon suggests that storytelling lays out the lives of protagonists for surveillance by those in control. Analysis of the seventeen narratives in the present corpus suggests that distribution of narrative roles in the Logan family is consistent with this pattern, as shown in Table 4.1. Table 4.1 demonstrates that the majority of stories are initiated by Meg, center on Meg's or the children's experiences, and are directed to William, establishing his power as inspector. Further, the majority of narratives Meg initiates (80%) center on her own distressing or otherwise unnerving experiences. By routinely narrating her anxiety-provoking experiences in this way, Meg establishes William in a position to (de)legitimize her emotions and actions as protagonist in the narrated scenario, as well as her present concern over this circumstance. Meg, perhaps like others who feel themselves to be "irrational," appears to seek validation (especially from "the world's nicest, most normal guy") for her framing of events and experiences, particularly those that remain unresolved. In the process, William obtains the authority to confer or withhold a judgment that Meg's feelings and behavior are rational, and his identity as "the world's nicest, most normal guy" is instantiated. Analyses of conarration in the Logan family illuminate these dynamics.

Table 4.1. Distribution of narrative roles during Logan family dinnertime interaction Member William

Meg

Initial teller

Primary recipient

4 (23.5%)

14(82%)

I (6%)

2 ( 1 2%)

7 (41%,)

1 0 (59%)

Principal protagonist

Beth

3 (17.5%)

3 (17%)

1 (6%)

Sean

0

0

4 (24%)

"Percentages do not add to 100 because two stories (12%) did not involve any family member as principal protagonist.

Constructing the Irrational Woman

89

deconstructing irrationality in tales of distressing experience The first of Meg's stories that will be considered here (Pit Bull Story #1) involves her unexpected encounter with two menacing pit bull dogs while paying a visit to her father-in-law (excerpts 1 a and Ib). In the telling Meg makes her actions, thoughts, and emotions accessible to other interlocutors, particularly to William, the primary recipient. Enduring mental activity Meg builds her narrative around her mental activity. She emphasizes her psychological responses to the presence of the dogs by routinely using mental verb constructions that highlight not only the content but also the very fact of her thoughts and emotions. These constructions convey the sense that she continues to feel distressed, in part by concern about the appropriateness of her anxiety. Similarly, Meg's subsequent narrative contributions (that is, her responses to responses) seek validation both of the content of her past reaction and of her ongoing emotional preoccupation with the encounter with the dogs.

(1 a)

Pit Bull Story #1 —Excerpt M = Meg (Mother) W = William (Father) B = Beth, age 11 S = Sean, age 6 S: looking at William W: [Well you don't- ((looking down at food)) M: [Turns body and gaze toward William And your dad the whole time was= W: = fdare trust em [looking down, leans away from Meg, toward Sean M: kind of saying good boy good boy and you know how your da:d is (.2) [He can make any animal (.) [Beth looks at Meg B: tame= M: =his friend. [But I couldn't help hut wonder [looks down at food, body oriented toward William what would happen [if \William does brisk vertical head nods still looking down, leaning away from Meg (.2) [you Tknow [Meg looks up at William (.4) Meg scratches top of head, then looks down at food W: They'd be in big trouble M: If he really did get into your dad's yard

90

IDENTITY AS INVENTION

(1 b) Pit Bull Story #1—Excerpt M: I got to thinking (.5) these [aren't PETS [looking down at food [looks up at William, does horizontal head shake [William looking at Meg (.6) [TNolbody would treat a [Tpct like that \Williamlooksdownatfood [Meg eyeflash at William [They're [strictly there to chew up [TanyJ-body [William looks up from food at Beth [Meg looks up at William [ William eyeflash at Meg, turns to Sean [who might come into their ya:rd [Sean wipes his mouth with his arm W: [Sure they are [William looks at Sean [Meg looks at Sean M: Wh-what [if that thing gets Tloo:se and Tki:lls somebody [Meg looks at William, who continues looking at Sean

When Meg uses forms such as But I couldn'? help but wonder what would happen if . . . and I got to thinking . . . , she is reporting not only what she is thinking and feeling but also that she is thinking and feeling. Mental verbs bring into focus Meg's consciousness of engaging in the activities of thinking and feeling and invite assurance that mental activity of this sort would be expected of a rational, normal person. In telling the story, Meg associates her thoughts and feelings with a particular past event: encountering the pit bulls. Yet the excerpts presented here also give the impression that her absorption with these thoughts and emotions continues through the moment of the telling; she remains distressed over the threat of the dogs. Meg constructs the objects of her thoughts and feelings in the present tense, depicting them as current, enduring concerns rather than issues contained in the past. She asks, for example, "what would happen . . ." rather than "what would have happened" and "what if that thing gets loose" rather than "what if that thing got loose." Through repeated use of these mental verb constructions, Meg indexes for her interlocutors that the predicament has not been resolved but continues to plague her present and imagined experience. In this way, Meg opens herself to others' assessments of the reasonableness of her past and present anxious worldview. Soliciting validation Meg not only opens herself to others' assessments of her thoughts and feelings but delivers the narrative in a way that actively elicits feedback on her thinking, in that utterances containing mental verb constructions arc characteristically delivered with hesitations that solicit responsiveness (C. Goodwin 1981; Sacks, Schcgloff, & Jefferson 1978). This behavior is not unique to Meg but is characteristic of women's speech styles more generally (see, e.g., Schiffrin 1987; Lakoff 1990; Tannen 1993; Watzlawick, Beavin, & Jackson 1967). Meg most actively solicits feedback from her

Constructing the Irrational Woman

91

husband—her body and eye gaze frequently orient toward him, requesting his participation as the preferred responder (C. Goodwin 1981, 1986)—both for the content of her thoughts and for being legitimately preoccupied at this point in the story. In these excerpts, however, William's validation of Meg is weak at best. In excerpt (la), for example, Meg presents statements conveying her thinking (wondering) about the pit bulls. Although William responds with brisk vertical head nods, his gaze remains fixed on his food, his body oriented away from Meg. After a short gap (0.2 seconds), Meg, who is looking at William, solicits additional feedback from him by asking, "you know?" William withholds feedback through a lengthier gap (0.4 seconds). In response to his silence—which signals disagreement or nonalignment (Davidson 1984; Pomerantz 1978)—Meg, still gazing at William, intensifies her display of confusion and request for input from him by scratching the top of her head. She then looks down at her food, at which point William says, "they'd be in big trouble." Here William does provide feedback, but in so doing empathizes not with Meg's past and ongoing distress over the situation but with his parents' predicament. Similarly, in excerpt (Ib), Meg presents four different statements conveying her thinking about pit bulls. Although William stares at Meg as she begins to relate her thoughts, he does not provide verbal feedback after her first predicate (these aren't PETS) nor after she looks straight at him and intensifies her message by shaking her head horizontally nor after a subsequent 0.6-second pause (which is a lengthy conversational invitation to take the floor). William continues to withhold substantive response during Meg's second thought (TNo-lbody would treat a ^pet like that), even though she flashes a glance at him. This prolonged lack of responsiveness can be taken as a strong sign of nonalignment. As Meg goes on to express a third thought (They're strictly there to chew up T'any-lbody who might come into their ya:rd), William looks up not at her but at his daughter Beth. He then glances at Meg momentarily en route to fixing his gaze on his son Sean. At this point, William finally responds verbally (Sure they are), but all the while his gaze remains fixed on Sean, signaling that his attention is divided and he is only partly available to her. Furthermore, this assessment is "downgraded" in that it does not match the intensity of Meg's portrayal. As Anita Pomerantz (1984) has pointed out, such responses undermine rather than ratify the assertion at hand. Continuing her bid for ratification, Meg goes on to express her fourth and most potent thought as a direct question to William, who is the object of her gaze; William, however, remains focused on Sean and never responds verbally or otherwise acknowledges the question. Escalating bids for responsiveness This conversational dynamic between Meg and William may perpetuate Meg's panicky thoughts and feelings associated with past or hypothetical events, which in turn may maintain her identity as an irrational agoraphobic. In response to silence or lowaffiliative, downgraded responses, tellers typically escalate the intensity of both the narrative situation and their bids for alignment from conarrators (Pomerantz 1984). Indeed, these excerpts illuminate how William's low-affiliative responses lead Meg to increase the frequency of her bids for responsiveness. In so doing, she also de-

92

IDENTITY AS INVENTION

scribes her experience in progressively more self-threatening and anxious terms. Meg intensifies the dangerousness of the situation and attempts to increase the relevance of the threat to other interlocutors by expanding its scope to include those copresent and by rendering the dogs a present-tense menace. For example, in (la), through the use of a hypothetical scenario, she casts the dogs as an enduring danger to anyone who visits William's parents, rather than confining the threat to her past experience: / couldn't help but wonder what would happen if. . . If he really did get into your dad's yard. Similarly, in (Ib) Meg describes the dogs as creatures who exist for the sole purpose of mauling "anybody" in the vicinity. Finally, Meg asks her halfattending husband to ponder with her the possibility that the one particularly threatening pit bull, described emphatically as a thing, could "get Tloo:se" and "Tki:ll somebody." As is the case in (la), these constructions increase the scope of the threat to include not just Meg but "anybody" and "somebody" and not only in the past but also in the present. This progression can be viewed as an attempt to engage William's participation by drawing him into the realm of the story. Indeed, good storytellers often attempt to involve interlocutors by dramatizing past events as if they are taking place in the here and now (Buhler [1934] 1990; Schiffrin 1981; Wolfson 1979). However, for Meg, it seems that these experiences are so vivid and pressing that they force their way into her current consciousness—rather than narratively controlling the past, Meg is herself controlled by it.4 These rhetorical strategies can also be brought to bear in discussing the coconstruction of irrationality. By casting the scenario with "anyone" and "someone" as protagonists (rather than herself), Meg entreats her interlocutors to put themselves in her shoes. At the same time, she constructs her actions and emotions as what would be expected of "anybody"; she attempts to fill her shoes with rational feet. In so doing, she attempts to legitimize her psychological responses as protagonist and her voice as narrator, but to no avail. Rather than constructing Meg as being like anyone or even someone else, the progressive escalation in this storytelling interaction seems to have the opposite outcome: It renders her unlike others in her midst. Meg's unratified attempts attest to the fact that she alone cannot construct herself as a rational woman—nor can any individual. Similarly, neither William nor anyone else can single-handedly confer irrationality. Social identities are an interactional achievement. Indeed, William's role as the "panoptical" primary recipient is also a collaborative construction by family members. Thus only through recurrent participation in such interactions are Meg's experiences and interpretations coconstructed as "irrational" and "abnormal," thereby constituting and perpetuating her agoraphobic identity. Coconstructing irrationality through alternative narrative roles Meg's identity as irrational is also coconstructed through family storytelling interactions that she does not initiate and that do not center on her distressing encounters as a protagonist. One such storytelling, again involving pit bulls, took place a few weeks after the first within the context of a discussion about the death of the Logan

Constructing the Irrational Woman

93

family's dog and whether or not to get another one. In (2a), William reports having seen an advertisement for part pit bull / rottweiler / German shepherd puppies and proposes that they adopt one as a family pet: (2a)

Pit Bull Story #2 W: I got to tell you what they were advertising in Comptonon Compton Boulevard on the way home ((looking straight ahead at Beth)) S: [looks up at William B: [looks up at William M: [What [looks up at William B: looks at Meg, and then at William W: We could ha:ve (.) a:: pit bull [and shepherd [Meg jerks backward, away from William and part rottl'weilcr puppy [Beth gets up from chair, looks out window HA HA HA ha ha M: [Oh yea:h [looks down at plate B: A pit bull [shepherd and -J.[rottweiler [walks out of the room, into the kitchen M: [Who needs a gun with a dog like that ((solemn tone))

Narratives have a family history Narrative interactions are constructed within the context of developing individual and collective histories. They are not isolated units but build on each other. In varying ways and degrees, each storytelling shapes those that precede and follow. Although stories may be related by theme, genre, or participant structure, each telling contributes to the ongoing construction of participants' identities (see also Wood, chapter 2, this volume). The second pit bull storytelling harks back to the first both thematically and interactionally. William's contributions to this story explicitly counter Meg's previous assessment of pit bulls and thus her authority to make valid, rational assessments. When William delivers his report he invokes and undermines Meg's prior evaluation of pit bulls as menaces to society by making light of the situation she deemed lethal. Moreover, in suggesting that they take such a dog into their home, William proposes that they enact the worst-case scenarios in Meg's prior account: In the first pit bull story (excerpts la, Ib), Meg posits a most-dreaded scenario in which the pit bull escapes and enters the family yard. In addition, William's suggestion that they adopt the pit bull mix as a pet explicitly opposes Meg's previously expressed thoughts: I got to thinking (.5) these aren't PETS. William's proposal thus directly refutes the legitimacy of Meg's narrative appraisal of the pit bulls she encountered outside her father-in-law's yard.

94

IDENTITY AS INVENTION

Bridging past and present narrative roles The second pit bull storytelling resembles the first in that both render Meg as having irrational thoughts and feelings about the dogs. In doing the telling as a joke, William constructs Meg as having been irrational or abnormal for being mired in anxious thinking about the dogs. In contrast to William's jocular tone, Meg's solemn assessment of the situation (Who needs a gun with a dog like that) reinforces this construction, affirming her ongoing concern about pit bulls. Further, the generalized form of her comment (Who) harks back to previous failed attempts to construct herself as rational or normal by aligning herself and her assessments with those of generic others. The third pit bull story is immediately touched off by the second. Following William's lead, Meg herself jokingly relates an incident in which a neighbor's motherin-law unwittingly locks herself in the backyard with a pit bull. In this sequence (2b) Meg constructs an account of a ridiculous if not irrational female protagonist whose distress over the pit bull is laughable: (2b)

Pit Bull Story #3—Excerpt M: [Joe and Charlotte have a pit bull [looking at William (.4)

He said his mother-in-law got locked out They have a Thouse (.2) they have athe door to the backyard locks automatically when it closes be(h)hind you And and ((laughingly)) she went out in the backyard She was visiting them and they didn't know that (.3) she was backin the backyard with the pit bull And all of a sudden she yells CHARLOTTE, CHARLOTTE OPEN THIS DOOR hee hec he he he ((laughs covering mouth with hand)) B: [WHAT [looking at Meg M: Joe's mother-in-law got locked out in the backyard with their pit bull and he said it would eat anybody B: SS-ss- HA ha ha M: [She didn't know [looks down at plate B: Oh really ((laughingly)) M: And they had the door that would W: [I'd be afraid to have it- to have that dog around the kids [looks at Meg B: What dog ((looks at William)) S: [Pit bull ((looks at Beth)) W: [The pit bull ((looks at Beth))

Constructing the Irrational Woman B: M: W: B: M: W: M:

95

[Oh. Across the Tstreet [looking at William No= =No OH AnyJoe and Charlotte's= =Any pit bull actually ((looking down at plate))

This third story is thematically linked to the first and second storytelling interactions. In this narrative, Meg locates their neighbor Joe's mother-in-law in the place in which she initially imagined herself and most feared being: in the yard of an unleashed pit bull. At the same time, Meg locates Joe's mother-in-law in the position she herself occupied in the course of William's joke-telling, which cast her distress as laughable. In telling a joke, Meg appears to be attempting to align with "normal" (that is, rational) William's lighthearted stance toward the situation. But paralleling the progression in the second pit bull storytelling, by the end Meg's rendering of the scenario is menacing, not laughable. As occurred over the course of the first and second storytellings, Meg's anxiety over the dogs becomes apparent as she portrays the pit bull as threatening not just to the protagonist but to everyone. For example, when she reiterates her account (following Beth's WHAT), Meg describes the pit bull as threatening, saying, "it would eat anybody." This portrayal converges with her depiction of the pit bulls in the first story—as animals that are "there to chew up Tanylbody who might come into their ya:rd." Similarly, as in the first storytelling, Meg's anxiety as coteller seems to escalate as her focus shifts from the particular pit bull in Joe and Charlotte's backyard to "any pit bull." Rather than being confined to a particular dog in a particular yard, the scope of the threat includes a multitude of dogs in any imaginable setting. The narrated circumstance is not contained in the past but pervades present and imagined time and space. As in the first and second stories, Meg seems unable to distance herself from her fear. Moreover, she is unsuccessful in her attempts to enlist family members' support for her position. Thus participation in these narrative interactions yields the outcome Meg most fears: being alone and experiencing inescapable, irrational fear of imminent danger. In the third pit bull story, this shift in tone is constructed in part by William's comment, I'd be afraid to have it- to have that dog around the kids. His remark seems to push Meg further into the expanding gulf of anxiety that envelops and overwhelms the particulars of this story. William's comment, however, frames the scenario as threatening to "the kids." This response does not validate the protagonist's—and by extension, Meg's—fears. Rather, William aligns himself with the protectors, the adults. This pattern is reminiscent of the first story, in which William responds to Meg's concern about what would happen if the pit bulls entered her father-in-law's yard—her location as protagonist in the story—by identifying with his parents' predicament: They'd be in big trouble. In such interactions William constructs fear of pit bulls as valid with respect to children and the elderly but not to people like himself. William thereby implies that he is not afraid for the female protagonists (the

96

IDENTITY AS INVENTION

mother-in-law, Meg) and to some extent that their fear is irrational and illegitimate. William further differentiates himself (and other normal people) from this irrational lot by restricting the scope of the threat to the initial story setting. Implications The present analysis of storytelling interactions in the Logan family suggests that members routinely assume narrative roles that construct Meg's agoraphobic identity: Meg frequently addresses stories—particularly stories of her own distressing encounters—to William, establishing him in a position to evaluate her rationality both as a protagonist acting, thinking, and feeling in the past and as a narrator acting, thinking, and feeling in the present. Meg's attempts to contextualize narrated attitudes and behaviors are rarely ratified by her husband ("the most normal guy in the world"). This dynamic propels an escalating cycle in which Meg senses herself to be irrational (that is, unlike normal others in her midst) and attempts to secure ratification for her distress by widening the scope of her narrated anxiety, which often leads William to withdraw further. Why is it that William does not enter deeply or for a sustained time into her narrative recollections and speculations? One possibility might be that he wishes to curtail panic. Interlocutors listening to Meg's accounts of her past anxieties may withhold feedback for an extended period of the narrative, perhaps to avoid validating her fears or aligning themselves with her anxiety. When Meg recounts prior anxious scenarios for her family, she brings her past anxieties into present collective consciousness. Members of her family, particularly William, may display only minimal responsiveness to such recountings to discourage Meg from continuing this process and perhaps from involving their children in experiencing the same fears. In other words, William's minimal displays of involvement may be attempts to shut down narrative emotionality before it gets out of hand and develops into full-blown panic, sweeping the family into submission. However, displays of minimal or no responsiveness by the primary recipient of the narrative—her spouse—lead to escalation, not curtailment, of expressed anxiety. For Meg, a persistent outcome of such interactions is a perception of herself as "less than," "crazy," "irrational," and otherwise abnormal, leading her to avoid places outside the home in which she fears experiencing such sensations and to identify herself "first and foremost as agoraphobic." Countering the view that psychopathology exists in the isolated minds of individuals, investigation of storytelling interactions in this family suggests that disorders are constructed in interaction. In contrast to decontextuali/cd research measures that categorize individuals according to self-reported traits, analysis of how people use language, gesture, and body orientation to negotiate the meaning of events illuminates identities and worldviews in the making. Studies of face-to-face interaction capture rather than mask complexities and undermine rather than perpetuate stereotypes (see also Goodwin, chapter 20, and Morgan, chapter 1, both this volume). This approach demonstrates that individuals do not simply "play out" preexisting roles and that identities are interactional achievements involving acts of resistance and unintended outcomes, achievements that cannot be divorced from a particular socio-

Constructing the Irrational Woman

97

historical context. Analysis of language in interaction locates convergences between ideologies of gender and of mental illness and illuminates how different speakers respond to these ideologies through practices that challenge and reproduce dominant beliefs. Finally, analysis of language in interaction reveals opportunities for reconfiguring identities and for assigning new meanings to gender. NOTES 1. See J. Maxwell Atkinson and John Heritage (1984) for transcription conventions. 2. This percentage is considerably higher than those found in previous investigations conducted in the United States, England, Germany, and Switzerland, which reported prevalence rates of agoraphobia ranging from 0.6 percent to 3.6 percent. The higher percentage reported in the epidemiological study is attributable to the fact that participants were asked about symptoms over the past year rather than at the time of the interview, as was the case in the other studies. 3. The statement also reveals the way in which research methodologies may perpetuate cultural biases, in this case a heterocentric view: Because most studies of agoraphobic persons' close relationships have involved agoraphobic women and their husbands, such work has promoted the belief that agoraphobia is a married women's syndrome. Yet according to epidemiological surveys, this is not the case (Eaton, Dryman, & Weissman 1991). 4. See Capps and Ochs (1995a, b) for further discussion of grammatical forms through which Meg's anxious past experiences overwhelm her present experience. REFERENCES

American Psychiatric Association (1995). Diagnostic and statistical manual of mental disorders. (4th ed.) Washington, DC: American Psychiatric Association. Arrindell, Willem A., & Paul M. G. Emmelkamp (1986). Marital adjustment, intimacy, and needs in female agoraphobics and their partners: A controlled study. British Journal of Psychiatry 149:592-602. Atkinson, J. Maxwell, & John Heritage (eds.) (1984). Structures of social action. Cambridge: Cambridge University Press. Bateson, Gregory (1972). Steps to an ecology of mind. New York: Ballantine Books. Breuer, Josef, & Sigmund Freud (1957). Studies on hysteria. Trans. James Strachey & Anna Freud. New York: Basic Books. Bruner, Jerome (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University Press. (1990). Acts of meaning. Cambridge, MA: Harvard University Press. Buhler, Karl (1990). Sprachtheorie: Die darstellungsfunktion der sprache. Jena: Gustav Fischer. (Original work published 1934) Capps, Lisa (1996). The psychological adjustment of children of agoraphobic parents. Ph.D. diss., University of California, Los Angeles. Capps, Lisa, & Elinor Ochs (1995a). Constructing panic: The discourse of agoraphobia. Cambridge, MA: Harvard University Press. (1995b). Out of place: Narrative insights into agoraphobia. Discourse Processes 19(3):407440. Capps, Lisa, Marian Sigman, Rhonda Sena, Barbara Henker, & Carol Whalen (1996). Fear, anxiety, and perceived control in children of agoraphobic parents. Journal of Child Psychology and Psychiatry 37(4):445-452. Chambless, Dianne L., & Jeanne Mason (1986). Sex, sex-role stereotyping and agoraphobia. Behaviour Research and Therapy 24(2):231—235.

98

IDENTITY AS INVENTION

Chesler, Phyllis (1972). Women and madness. New York: Avon. Davidson, Judy (1984). Subsequent versions of invitations, offers, requests, and proposals dealing with potential or aetual rejection. In J. Maxwell Atkinson & John Heritage (eds.), Structures of social action. Cambridge: Cambridge University Press, 102-128. DiNardo, Peter A., Gerald T. O'Brien, David H. Barlow, Maria T. Waddell, & Edward B. Blanchard (1983). Reliability of DSM-III anxiety disorder eategories using a new structured interview. Archives of General Psychiatry 40:1070-1074. Duranti, Alessandro (1986). The audience as co-author: Introduction. Text 6(3):239-247. Eaton, William W., Amy Dryman, & Myrna M. Weissman (1991). Panic and phobia. In Lee Robins & Darrel A. Regier (eds.), Psychiatric disorders in America: The American Epidemiological Catchment Area Study. New York: Free Press, 155-179. Feldman, Carol F. (1989). Monologue as problem-solving narrative. In Katherine Nelson (ed.), Narratives from the crib. Cambridge, MA: Harvard University Press, 98-122. Foa, Edna B., Gail Steketee, & Mark C. Young (1984). Agoraphobia: Phenomenological aspects, associated characteristics, and theoretical considerations. Clinical Psychology Review 4:43 1 — 457. Fodor, Iris (1974). The phobic syndrome in women: Implications for treatment. In Violet Franks & Vasanti Burtle (eds.), Women in therapy. New York: Brunner/Mazel, 132-168. Foucault, Michel (1965). Madness and civilization. New York: Vintage. (1979). Discipline and punish: The birth of the prison. New York: Random House. Goffman, Erving (1959). The presentation of self in everyday life. Garden City, NY: Doubleday. Goldstein, Alan J., & Dianne L. Chambless (1978). A rcanalysis of agoraphobia. Behavior Therapy 9:47-59. Goodwin, Charles (1981). Conversational organization: Interaction between speakers and hearers. New York: Academic Press. (1986). Audience diversity, participation and interpretation. Text 6(3):283-316. Goodwin, Charles, & Alessandro Duranti (1992). Rethinking context: An introduction. In Alessandro Duranti & Charles Goodwin (eds.), Rethinking context. Cambridge: Cambridge University Press, 1-42. Goodwin, Charles, & Marjorie Flarness Goodwin (1992). Assessments and the construction of context. In Alessandro Duranti & Charles Goodwin (eds.), Rethinking context. Cambridge: Cambridge University Press, 147-190. Goodwin, Marjorie Harness (1991). Retellings, pretellings and hypothetical stories. In Lea Laitinen, Pirkko Nuolijarvi, & Mirja Saari (eds.), Leikkauspisle. Helsinki: Suomalaisen Kirjallisuuden Seura, 43-58. Hafner, R. Julian (1982). The marital context of the agoraphobic syndrome. In Dianne L. Chambless & Alan J. Goldstein (eds.), Agoraphobia: Multiple perspectives on theory and treatment. New York: Wiley, 143-170. Heath, Shirley Brice (1983). Ways with words: Language, life, and work in communities and classrooms. Cambridge: Cambridge University Press. Jacoby, Sally, & Elinor Ochs (1995). Co-construction: An introduction. Research on Language and Social Interaction 28(3):171-183. Kleiner, Liliana, & W. L. Marshall (1987). The role of interpersonal problems in the development of agoraphobia with panic attacks. Journal of Anxiety Disorders 1:313—323. Labov, William, & Joshua Waletzky (1968). Narrative analysis. In William Labov, Paul Cohen, Clarence Robins, & John Lewis (eds.), A study of the non-standard English of Negro and Puerto Rican speakers in New York City. New York: Columbia University, 286-338. Lakoff, Robin (1990). Talking power: The politics of language in daily life. New York: Basic Books.

Constructing the Irrational Woman

99

Lakoff, Robin, & James Coyne (1993). Father knows best: The use and abuse of power in Freud's case of Dora. New York: Teachers College Press. Marks, Isaac M., & E. R. Herst (1970). A survey of 1,200 agoraphobics in Britain. Social Psychiatry 5:16-24. Miller, Peggy J., Randolph Potts, Heidi Fung, Lisa Hoogstra, & Judy Mintz (1990). Narrative practices and the social construction of self in childhood. American Ethnologist 17:292311. Ochs, Elinor (1988). Culture and language development: Language acquisition and language socialization in a Samoan village. Cambridge: Cambridge University Press. (1993). Stories that step into the future. In Douglas Biber & Edward Finegan (eds.), Sociolinguistic perspectives on register. Oxford: Oxford University Press, 106-135. Ochs, Elinor, & Bambi Schieffelin (1984). Language acquisition and socialization: Three developmental stories. In Richard A. Schweder & Robert A. LeVine (eds.), Culture theory: Essays on mind, self, and emotion. Cambridge: Cambridge University Press, 276-320. Ochs, Elinor, & Carolyn Taylor (1992a). Family narrative as political activity. Discourse and Society 3(3):301-340. (1992b). Mothers' role in the everyday reconstruction of "Father knows best." In Kira Hall, Mary Bucholtz, & Birch Moonwomon (eds.), Locating power: Proceedings of the Second Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 447—462. (1995). The "Father knows best" dynamic in dinnertime narratives. In Kira Hall & Mary

Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 97-120. Ochs, Elinor, Carolyn Taylor, Diria Rudolph, & Ruth Smith (1992). Story-telling as a theorybuilding activity. Discourse Processes 15(l):37-72. Pomerantz, Anita (1978). Compliment responses: Notes on the co-operation of multiple constraints. In Jim Schenkein (ed.), Studies in the organization of conversational interaction. New York: Academic Press, 79-112. (1984). Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In J. Maxwell Atkinson & John Heritage (eds.), Structures of social action. Cambridge: Cambridge University Press, 57-101. Sacks, Harvey, Emanuel Schegloff, & Gail Jefferson (1978). A simplest systematics for the organization of turn taking in conversation. In Jim Schenkein (ed.), Studies in the organization of conversational interaction. New York: Academic Press, 7-55. Sass, Louis (1992). The paradoxes of delusion. Ithaca, NY: Cornell University Press. Schieffelin, Bambi (1990). The give and take of everyday life: Language socialization of Kaluli children. Cambridge: Cambridge University Press. Schiffrin, Deborah (1981). Tense variation in narrative. Language 57(l):45-62. (1987). Discourse markers. Cambridge: Cambridge University Press. Szasz, Thomas S. (1974). The myth of mental illness. New York: HarperCollins. Tannen, Deborah (1993). The relativity of linguistic strategies: Rethinking power and solidarity in gender and dominance. In Deborah Tannen (ed.), Gender and conversational interaction. New York: Oxford University Press, 165-188. Watzlawick, Paul, Janet Beavin, & Don Jackson (1967). Pragmatics and human communication: A study of interactional patterns, pathologies, and paradoxes. New York: Norton. Weissman, Myrna (1993). Family genetic studies of panic disorder. Journal of Psychiatric Research 27(l):69-78. Wenegrat, Brant (1995). Illness and power: Women's mental disorders and the battle between the. sexes. New York: New York University Press.

100

IDENTITY AS INVENTION

White, Hayden (1980). The value of narrativity in the representation of reality. In W. J. T. Mitchell (ed.), On narrative. Chicago: University of Chicago Press, 1-23. Wolfe, Barry E. (1984). Gender imperatives, separation anxiety, and agoraphobia in women. Integrative Psychiatry 2(2):57-61. Wolfson, Nessa (1979). The conversational historical present alternation. Language 55(19):168182.

5

SARA TRECHTER

Contextualizing the Exotic Few Gender Dichotomies in Lakhota The characteristics between male and female language are very peculiar. I think, before assuming the origin of the language as due to admixture of tribes, it might be well to consider whether the mannerism of speech of different social groups may not be a sufficient explanation. Take, for instance, our students' slang, which is only used in conversation in college, but which the students readily drop when talking to other people. —Letter from Franz Boas to Edward Sapir concerning gendered language in Yana, August 28, 1907

istorically, the field of gender and language has been discursively framed in terms of oppositions: women's and men's speech (Haas 1944), exclusive and preferential gender (Bodine 1975), community and contest (Johnstone 1993). Recent feminist theory has questioned the biocultural presuppositions that underlie gender oppositions, locating the deconstruction of gender firmly in social interaction, where simple oppositions are difficult to maintain. Nevertheless, as Janet Bing and Victoria Bergvall (1996) have noted, language and gender researchers have sometimes been content to refine their contestation of dichotomous thinking in social terms rather than rid the field of such constructs entirely. Typically, academics no longer speak in terms of the first pseudodichotomy listed above, women's and men's language, but they sometimes focus on differential strategies that women and men practice as members of supposedly separate communities. Inevitably, any dichotomous framing either obscures or assimilates the diverse identities of those who cannot be recognized within that frame. Marcyliena Morgan (chapter 1, this volume) points out, for instance, that the way sociolinguists have cast the dichotomy for speakers of African American English has left out vital populations. In sociolinguists' effort to see the difference between European American and African American speakers, the class and gender diversity of both populations is ignored. One might say that the ability to name difference at all is a first step in the recognition of diversity and therefore of the other as separate from self. However, Judith 101

102

IDENTITY AS INVENTION

Butler (1997) does not support such one-sided naming in her analysis of hate speech, arguing that it is through a dialogic process of addressing (or even name-calling) that subjectivity and agency are achieved. Moreover, Butler argues that adroit response to insistent categorization can subvert the meanings of an oft-repeated appellation. The appropriate comeback is not always at hand or easy to assert within sociohistorical confines of racism, sexism, or homophobia, but many of the chapters in this volume offer strategies to confront an unacceptable appellation. Caitlin Hines, for example, exposes the historical and ideological roots of the "woman as dessert" metaphor to weaken its deceptive force. And Morgan, Marjorie Goodwin, and Norma MendozaDenton engage in dialogue with the limiting categories of prior scholarship by contextually foregrounding the actual voices and emergent identities of African American women and Latina girls, respectively. In this chapter, I explore the nature of gendered language in Lakhota in response to previous researchers' claims that some Native American languages such as Lakhota code gender rigidly through grammar or phonology (Bodine 1975; Haas 1944; Sapir 1949). I employ two strategic responses to earlier work. To understand the linguistic resources that Lakhota speakers draw on to establish gendered identities, I accentuate the various positions expressed in local contexts and emphasize how cultural scripts become associated with gendered behavior in specific interactions. The ways in which Lakhota speakers assert their gender through speech acts cannot stand without an ideological corollary, however. The Lakhota also speak about and frame their own gendered language use. Like the analyses of linguists, this metapragmatic talk is usually characterized by binaries. Rather than deny the existence of gender abstractions by further pointing to contextual and interactional fact, my second strategy is to expose the roots of native gender constructs in contextualized language by probing how their insistent force is silently invoked even as speakers confront them. The context of gender and language in Native America The association of phonological and morphological gendered differences with Native American languages is largely the result of two brief articles published by Mary Haas (1944) and her mentor, Edward Sapir (1949).' In six pages, Sapir outlined morphological and pronunciation differences in man-to-man speech inYana society. Haas's seven-page article, "Men's and Women's Speech in Koasati," detailed the phonological variation present in Oklahoma Koasati, wherein speakers systematically pronounced verbs differently depending on their sexual categorization as female or male. Haas's major focus was interesting phonological derivations, which she illustrated through verbal paradigms. She also cited several unrelated languages—Thai, Chukchee, Biloxi (Siouan), Tunica, and Yana—to show that the supposedly rare phenomenon of distinctive women's and men's speech was not really so unusual. Two years later Regina Flannery (1946) published a three-page account of a similar pronunciation distinction in Atsina (Gros Vcntre), an Arapaho dialect spoken in Montana. Like those before her she focused on linguistic rather than social contexts. Brief examples of the gender paradigms emphasized in these languages are given in (1).

Contextualizing the Exotic Few (1)

103

"Exclusively" gendered forms in Native American languages2 a. Koasati (Kimbail 1987) female male alohlq alolo:J' 'He can drive.' Akpq akpo:f 'I do not eat.' Ocintq ocmto:f 'You can come.' b. Atsina/Gros Ventre (Flannery 1946) female male [k] or [ky] [c] ikenibik icenibic 'his gum' c. Yana (Sapir 1929) female male ya yana 'person' nisat h nisat h i 'it is said he goes' p h at h p h adi 'place'

The Native American gender articles must be understood within the social and academic context in which they were produced. Coercive government policies mandating linguistic and social assimilation of Native Americans had already had a debilitating effect on Native American linguistic communities (see Dobkins, chapter 9, this volume). Anticipating the imminent death of Native American languages, linguistic anthropologists' stated purpose was to gather as much cross-linguistic information as possible before it was too late. Detailed social analysis of language use was difficult when the researcher was still struggling with basic grammar and lexicon. Consequently, the brief gender articles should be taken as contributions to a linguistic data pool rather than sociolinguistic analyses of gender. Of course, at the time of Boas and Sapir's correspondence in the early part of the twentieth century, anthropological linguists were already aware of sentence-final particles that gendered the identity of the speaker or addressee in the Siouan languages (such as Lakhota) and in many of the languages mentioned by Haas. Nonetheless, Haas's work was an important early development in the field of gender and language. She published her analysis of Koasati and a typology of speaker-addressee gender indicators in the journal Language, making the data available to linguists working outside of the (Native) Americanist tradition and to subsequent language and gender researchers.3 Haas's groundbreaking work is still influential, but perhaps not in the way she would have intended as a descriptive linguist who worked closely with Native Americans. Koasati (Muskogean), Yana (Hokan), and Atsina (Algonquian) are still cited in the literature and introductory linguistic textbooks as possessing exclusively gendered speech in which there is an isomorphic relationship between a form and the sex of the speaker or addressee (Bodine 1975; Bonvillain 1993; Finegan 1999; O'Grady, Dobrovolsky, & Aronoff 1997). These "gender-exclusive" languages are typically juxtaposed to those that are "gender-preferential" (or variable). And writers often exoticize them by contrasting the socially "dramatic" gender distinctions in Koasati with the distinctions in English (Finegan 1999:407). The languages "in which men and women always use linguistic alternatives appropriate to their own gender"

104

IDENTITY AS INVENTION

are invariably those cited in Haas's (1944) article (Bonvillain 1993:215; my emphasis). The brief treatment of language and gender in the textbook Contemporary Linguistics most clearly exemplifies this appropriation, and so I will quote it at length: Gender-exclusive differentiation refers to the radically different speech varieties used by men and women in particular societies. In these societies, a woman or man may, except in special circumstances, not be allowed to speak the variety of the other gender. . . . A society in which this is the norm is typically one in which the roles assigned the genders are rigid, and in which there is little social change. Gender-variable differentiation is much more common in the languages of the world than is gender exclusivity. This phenomenon is reflected in the relative frequency with which men and women use the same lexical items or other linguistic features. (O'Grady et al. 1997:518-519; my emphases)''

Combined with the authors' placement of Koasati examples under the "exclusive" heading and English examples under the "variable" heading, it is not difficult to read this social reworking of Haas's text as a manifestation of linguistic exoticism along the following lines: People who have gendered linguistic forms are radically different from European Americans; they are people who actively restrict the speech of others in their rigid societies, whereas we have the choice to prefer certain linguistic variables over others in our free, modern society.5 Although a few Native American languages are reported over and over as possessing gender-exclusive systems, there is considerable doubt that this has ever been the case, and it is more than likely that gender-exclusive systems do not exist in these languages. Even Haas observed that the Koasati take on the voices of other genders when quoting and in folktales. She also maintained that younger women were switching to the "male" forms. Regina Flannery (1946) reported that when men use women's speech in Atsina they are regarded as effeminate, which clearly implies that some men must have been using "women's" forms. Perhaps these are the "special circumstances" William O'Grady and his colleagues refer to above, where deviant gender behavior is allowed or restricted by a society. Yet masking all difference between people under the gender binary is problematic for the reasons mentioned above and in light of recent language and gender research and queer linguistic theory (Bergvall, Bing, & Freed 1996; Hall & Bucholtz 1995; Livia & Hall 1997). For instance, Rusty Barrett's discussion (chapter 16, this volume) of African American drag queens illustrates that taking on the pronunciation styles of another in the context of a performance does not always signify only one meaning, and it may in fact signify that gender is not binary. Native American men's use of "women's language" may be similarly polyscmous. Without sociocultural detail it is impossible to know in what contexts, or even if, effeminacy is a cultural taboo for male speakers in different Native American cultures. Many Native American peoples such as the Lakhota and the Dineh (Navajo) traditionally recognize more than two genders.6 Finally, even if crossing an imaginary gender line is a cultural taboo, such linguistic crossing may have a healthy linguistic existence. Kira Hall and Veronica O'Donovan (1996) have shown in their analysis of the referential language of hijras (male eunuchs) in India that referring to the self or other hijras

ContextuaUzing the Exotic Few

105

with both female and male gendered forms is systemic and contextually meaningful. Given these perspectives, one might be tempted to regard the idea of gender-exclusive linguistic features as dubious. In fact, several researchers have been puzzled on finding that the speakers they worked with had no clear notion of "women's" or "men's" ways of speaking. The Koasati people whom Geoffrey Kimball (1987) consulted did not use "male" forms of speech at all, even though Haas stated that women (and men) of the younger generation were adopting the "male" pronunciation. Kimball (1987, 1990) also discovered examples of women in the early part of this century using the "male" indicative/imperative -/'marker (which Haas described as a phonological rather than morphological phenomenon). One of these women was a famous doctor, and the other was the daughter and wife of a chief. In a move reminiscent of William O'Barr and Beryl Atkins's (1980) reanalysis of Robin Lakoff's (1975) women's language as powerless language, Kimball reinterpreted the use of the Koasati -/as an indication of status or respect rather than gender. In the most recent account of gender in Atsina, Allan Taylor (1982) discovered that some men, if they display different styles at all, use "female" pronunciation features when speaking to children or outsiders, employing what they perceive to be an "easier" style of pronunciation in order to communicate. Taylor speculates that because Atsina children are traditionally cared for by women, they first learn the "female" pronunciation through caretaker language. Subsequently, some male children adopt "male" pronunciation, given enough exposure and overt socialization. Finally, there is doubt about the function of male-tomale speech in Yana, for Herbert Luthin (1991) found that men sometimes used "male" speech when speaking to women in formal contexts. Luthin concluded that Sapir's gender differences were in fact register differences. Although there has been some debate over Kimball's reanalysis of the -/in Koasati as a status marker (see Saville-Troike 1988), the doubts expressed concerning the validity or completeness of previous work have not reached the ears of the linguistic community as a whole. Just as anthropological linguists were aware of gendered language associations at the beginning of the twentieth century, Americanists now are aware that these associations lack definition, but that in most cases nothing can be done. A thorough sociolinguistic (re)analysis requires speakers who can interact in different social situations or records that provide different genres of texts to ascertain if linguistic differences correlate more nearly with register, gender, or both. A reexamination of the kind of gender-categorizing systems that occur in Koasati (fewer than 200 speakers), Atsina (fewer than 10 speakers), or Yana (no speakers) would be impossible for several reasons, all of which concern the loss of available speakers and consequently the sociolinguistic contexts of their daily interaction. To begin with, in languages that are becoming obsolescent, the use of phonological, lexical, and grammatical indicators of distinct social categories is often lost through acculturation because the society no longer retains social autonomy. It is also difficult to ascertain the extent to which dialect variation plays a role as researchers question the validity of gendered speech accounts of the past. Claims for the existence of gendered forms vary a great deal from dialect to dialect, which may cause misinterpretation in languages for which there is little dialect information from an earlier

106

IDENTITY AS INVENTION

period. Kimball (1990) and Muriel Saville-Troike (1988), for instance, disagree about gendered speech in Koasati partially because they worked with speakers in different locations. Furthermore, speakers who acquire dying languages often do so under restricted circumstances. For example, men may learn gendered forms from female speakers, such as grandmothers, who use primarily the female forms. With so few speakers it is difficult to ascertain if gender usage is related to conversational context, register, or status of participants when because of sheer lack of numbers, there are increasingly few contexts in which native speakers of the language converse. Finally, examination of the historical records reveals little, for much of the elicited speech was in the form of traditional literature or historical narrative in which speech was often idealized according to stereotypical norms. Complete reliance on records is therefore inadvisable.

The Lakhota context Of all the Native American languages that have traditionally been cited as having "exclusive" gender, Lakhota is one of the few that still has a significant number of speakers. It is difficult to judge the actual number, but Dale Kinkadc (1991) estimates that there are approximately 15,000 speakers of the five Siouan dialects remaining in Canada and the United States. Lakhota is one of these dialects. According to native speakers and linguists, Lakhota possesses morphological indicators of the sex of the speaker, which typically affix to the last verb in a sentence (Boas & Deloria 1941; Rood & Taylor 1997). Not only is gender still viable in Lakhota, but the historical records are numerous and particularly good because native speaker and linguist Ella Deloria had the foresight to write down conversations in which she took part, giving minute descriptions of participant relationships and overall context. In this study, I have drawn on both Deloria's sources and my own recordings of contemporary conversations in order to account for the relationship between stereotypical representations of gendered speech and its actual function.

The Deloria texts The Deloria texts are a rich collection of different genres, including impromptu conversations, autobiographical accounts, historical narratives, folktales, aphorisms, puns, political speeches, and songs. Deloria could not tape-record speakers. In some cases, she transcribed the Lakhota simultaneously if the speaker was able to hold her or his train of thought while she typed (Deloria 1937?). But because of ficldwork constraints, she usually had to recall the "voice," the personal style and individuality of the narrator, after she returned home. Although she was meticulous about capturing dialect and individual variation, she did not indicate the "slips"—false starts or silent pauses that exist in informal speech. She did try to capture each individual's pitch, intonation, and voice quality, such as whispering, however (1937b:349). Whether as a native speaker Deloria regarded the use of morphology appropriate to the other sex as a "slip" we shall never know. In any case, her ethnographically detailed descriptions

Contextualizing the Exotic Few

107

provide an invaluable souree of contextual soeiolinguistie information. Deloria's native perspective combined with her excellent skills as a linguist give life to her texts in ways that outsiders with primarily scientific linguistic goals could not.

Fieldwork context To supplement the textual data and to understand more about the social-interactional use of gender particles in Lakhota, T conducted field-work at Pine Ridge Indian Reservation in 1993. I initially chose this site for the purpose of comparison; Deloria had collected most of her texts in and around Pine Ridge. The current and historical sociopolitical exigencies at Pine Ridge, however, affected my fieldwork and data dramatically (as illustrated later in example 5). Shannon County, South Dakota, the location of Pine Ridge Indian Reservation, is the poorest county in the United States: The per capita income ($3,417) is the lowest in the nation, and unemployment is approximately 80 percent (Carlson 1997:8). Inhabitants liken their home to a "ThirdWorld country in the middle of the United States" and to "the inner city." It is therefore ironic that Pine Ridge has a history of being overrun every summer by people eager to delve into the Lakhota way of life, such as anthropologists. Likewise, Europeans and Americans wanting to live like "real" Indians as they quest for spiritual enlightenment have recently joined the seekers of the authentic Native American "experience" (Powers 1994). Although imitation may be seen as a form of flattery where all other things are equal, many Pine Ridge residents regard the summer influx as an attempt to steal their culture, religion, and language. I will not detail my problematic position as a European American linguist in the continuing colonialization of Lakhota culture as outlined here; needless to say, it is precarious (see Trechter 1998). Finding consultants, taping, and translating conversations were accomplished with the help of Eli James, an Oglala/Sichangu Sioux, and his network of family relations and friends. All the consultants were between the ages of 40 and 80, largely because older people were more certain of their abilities to conduct conversations in Lakhota. The conversations ranged from short snatches of speech (unrecorded but transcribed) to 20-minute interchanges to 2-hour "chewing-the-fat" conversations around the kitchen table. All the participants were aware that they were being recorded and knew that I was generally interested in gender and language.

Gender in Lakhota Lakhota speakers who were unaware of my interest in gender have often told me that women and men end their sentences in Lakhota differently. My Pine Ridge "permission to record" agreement contained a statement alluding to my interest in gendered speech styles. On reading this agreement, potential consultants directly addressed what they perceived to be my research question. Their explanation took the form of metapragmatic judgments or maxims about appropriate speech, which were the same as those offered in previous conversations: "Men say yo and women

108

IDENTITY AS INVENTION

say ye" or "Men say lo and women say le." Such maxims are significant to linguistically gendered identity. Through overt ideology, they allude to possibilities of gendering talk while simultaneously restricting participation through the citation of opposite and complementary speech behaviors. The specific forms cited in these maxims are particularly interesting for two reasons: first, the forms associate gender directly with speech acts, and second, even in stereotypic representations of actual speech such as folktales, the maxims do not hold true. Although certain interjections are gendered, Lakhota speakers associate gender most often with illocutionary or affective force. These sentence-final indicators of speech acts enable the speaker to take a stance or highlight her or his presence in relation to some proposition. (2a)

osnf? 'It's cold.'

(2b)

osnf yea 'It's cold assertion.'

For instance, in explaining to me the differences of meaning between (2a) and (2b), James emphasized that there is no propositional difference. He went on to clarify through the following example: "In the morning, my stepfather might say 'oxm?'— it's cold. It's often cold here in the winter. Maybe he heard the weatherman. Maybe he's been outside. But when he comes in the room (demonstrating, touching finger to mouth and holding it up), he says, 'Ah, osniye.'" In James's hypothetical example, the speaker indicates through the use of ye that he is the immediate source of the proposition "It is cold"; the use of an illocutionary force marker therefore subtly focuses the speaker in relation to his speech act. Hence, the fact that some speech acts are more or less evocative of gendered selves is in no way remarkable if we see that indicating the presence of the speaker's identity leads to her or his contextual embodiment. Characterization of the speaker's identity through forms which emphasize her or his bodily presence, affect, stance, or authority occurs in many languages of the world, such as Japanese (Ochs 1992) and the Muskogean languages (Karen Booker, personal communication). Butler (1997:152) likewise argues that speech acts imply the body because in performance the body is "the rhetorical instrument of expression." In Lakhota overt speech-act markers instantiate the "presence of the body" in relation to an expressed proposition. They therefore implicate gendered readings. Keith Walters (chapter 10, this volume) questions whether it is in such contextual embodiment that a person's multiple identities necessarily converge, but speakers' reliance on gender as the defining category that underlies Lakhota speech acts assumes such convergence. Linguists' and native speakers' emphasis on performative differences in the way women and men create and characterize a single gendered identity erases much of the subtlety of the act as described by Walters and leads to the binary oppositions summarized in table 5.1.7 However, in a cross-genre examination of the different speech acts, in which I focused on the performative act rather than privileging the sex as the category of analysis, a very different picture emerged

109

Contextualizing the Exotic Few Table 5.1. Lakhota clitics, by gender and speech act Women

Men

Formal question

hvjwe (obsolete)

huwo

Imperative

yea

yo lo

lllocutionary/Affective

force

Opinion/emphasis

le (archaic), yea

Emphatic statement

kj'to

Entreaty

na

yee

Surprise/opinion

m;{

wg

(Trechter, forthcoming). A summary of these results is given in table 5.2. The gendered forms are no longer complementary and exclusionary in the same ways. Only three forms are used "exclusively" by men (lo, yo, hitwo) in the Deloria texts, and two are used "exclusively" by women (na, mq). In fact, Ella Deloria (1937a: 118) astutely noted that "Ye is feminine, only in a negative sense. A man adds lo, to ye; a woman does not say lo, but stops with ye. Both say ye." Although this refocusing shows us that there are no complementary binary oppositions in Lakhota, it nevertheless supports native speakers' metapragmatic judgments that certain forms are exclusively gendered. In the remainder of this chapter, I demonstrate how claims of exclusive gender may be accurate in terms of prototypical categories of usage but that in actual speech events speakers manipulate context and their roles within it by choosing which gendered clitics to use and whether to use them at all. We can further our understanding of the gender deixis system in Lakhota by adapting the notions of frame and framework from William Hanks's work on spatial deixis into the social realm. For Hanks, a frame "denote[sj a set of lexical items whose members correspond to different parts of a ... conceptual whole" (1993:128). Any model of the gender system in Lakhota represents the prototypical associations of meaning and use for some hypothetical context. In this sense, it is a schematic frame. For example, to understand the speech act behind yea and yo, one must at least understand that they work within a cultural frame of gender and as imperatives in distinc-

Table 5.2. Lakhota clitics, revised Illocutionary force

Women

Men

Formal question

huwo

Imperative

yo

Opinion/Emphasis Entreaty

lo

na

More used by women

More used by men

yea

yee kjio

Emphatic statement Surprise/Opinion

Both

1114

Wi(

1 10

IDENTITY AS INVENTION

tion from each other, yea typically being used for a woman's imperative and yo a man's. A framework, by contrast, is "the immediate social field of space and time perception, orientation, and participant engagement in acts of reference" (1993:127). A frame therefore contains structural aspects of meaning, which are conventional and fairly fixed, but a framework deals with a specific instance of language use. The framework is a local, variable production in space and time and contains the participants' orientation with respect to the social setting and the particular meanings they are constructing. Within specific frameworks, the meaning produced by using gendered performatives constructs a speaker with respect to the speech event. 8 The gendered particles of Lakhota display variability of usage that depends on a number of factors: age, gender, personality, and authority. I shall first illustrate how underlying meanings of these gendered forms emerge from stereotypical contexts in which the meaning of the stance of the participant is fairly standard. Although these contexts represent stereotypic meanings, they are nevertheless frameworks because they orient a speaker to a speech event. Second, I will demonstrate how the prototypical associations of the meaning of a gendered particle vary from the norm in nonstereotypical frameworks while still retaining the underlying meanings. Stereotypic meanings in context Stereotypic speech is well represented in folktales, in which the speaker attributes gendered speech to a character in the narrative to illustrate her or his role in the plot. Because the gendered morphology is indicative of affect as well as illocutionary force, these markers function to represent the "voices" and therefore the characters in a story. In effect, they function to establish the heteroglossia of any discourse (Bakhtin 1981). For instance, le, which is conventionally associated with the speech of a woman, also carries a sort of maternal or nurturant quality in its prototypical usage when a woman is speaking to children or someone she cares for, as in examples (3) and (4). In example (3), the woman is portrayed as overly concerned for her husband's welfare. To represent her as sweet-talking him, the narrator uses le (with falling intonation) to express her position of exaggerated womanly concern. Example (4) emphasizes a grandmother's concern and relief on finding her twin grandsons safe after a thipi (dwelling) fire. Here the use of le and the diminutive form -la on the verb indicate the characters' connection and the speaker's regard for the naughty twins. (3)

wanituk h a yele. 'How tired you must be £.' (Deloria 1937b)

(4)

Hinii, ufka mic h ikfi toh;ni ofotamakit'apila tk h a yele? 'Well, did my poor little sons almost choke to death from the smoke f?' (Deloria 1932)

In contrast, lo, conventionally associated with a man's statement or opinion, implies a certain degree of masculine authority, especially when used by an elder male delivering a final opinion. The speech in (5) contains a large number of such statements.

Contextualizing the Exotic Few

111

The 80-year-old speaker, Luke, in this example had returned to his home to find me outside recording a story that his son (Pete) was telling. After asking his daughter (Mary) and James if I was Lakhota, and establishing that I was in fact waficu 'white' and a linguist, he began to talk very pointedly to me. This excerpt is directed toward me as the addressee, but Luke is responding as well to another conversation with outsiders that he has just arrived from. After setting up this context, he begins by imitating a white speaker's obnoxious question in English in a high-pitched voice. (5)

Modern authoritative male speech. 1 Luke: . . . "We lo:ve your culture; why don't you practice it?" Hiya, ukiyc ukit h awapi c h a 2 iyutok h apifni ye 16. Ho he, Ho he, yake kj he e ye 16. 3 Mary: OhEjtii ujak ll otapi c h a [j: 4 Luke: Taku] a:taya yeye hayapi wajte ma?i{ we 16. 5 Ha ki le ntj we ki lena lak h ota (speaker pinches the skin of his forearm and holds it up). (James and Mary begin to carry on another conversation simultaneously.) 6 Na. nif a:taya lak h ota. hayapi m± lena k h o lak h ota nq. owi tha.kikiyQ nu eyaj ( ) 7 n
srA8€M 16 Graciela: =[Hhhnh:: 17 Marisol: =[ No, si digo algo, no?= =[No:, I do say something, no?= «*• M X A =>Aha, aha, aha. G 34 All except Graciela, joining in gradually: GRA:-CI:-E:-LA:::: •a- A,M,L,Y,N -> G 35 All including Graciela: HhhhhHhhhh:: ((hugging)) REFERENCES

California Style Collective (1993). Variation and personal/group style. Paper presented at the annual New Ways of Analyzing Variation Conference, Ottawa. Camarillo, Albert (1985). Chicanos in California: A history of Mexican Americans in California. San Francisco: Boyd & Eraser. Canfield, D. L. (1981). Spanish pronunciation in the Americas. Chicago: University of Chicago Press. Davidson, Judy (1984). Subsequent versions of invitations, offers, requests, and proposals dealing with potential or actual rejection. In J. Maxwell Atkinson & John Heritage (eds.), Structures of social action: Studies in conversation analysis. Cambridge: Cambridge University Press, 102-128. Eckert, Penelope (1993). Cooperative competition in adolescent "girl talk." In Deborah Tannen (ed.), Gender and conversational interaction. New York: Oxford University Press, 32— 61. Eckert, Penelope, & Sally McConnell-Ginet (1992). Communities of practice: Where language, gender, and power all live. In Kira Hall, Mary Bucholtz, & Birch Moonwomon (eds.), Locating power: Proceedings of the Second Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 89-99. Eraser, Brace (1992). Types of English discourse markers. Ada Linguistica Hungarica 38(1—4): 19— 33. Goodwin, Charles (1979). The interactive construction of a sentence in natural conversation. In George Psathas (ed.), Everyday language: Studies in ethnomethodology. New York: Irvington, 97-121. (1995). Co-construction in conversations with an aphasic man. Paper presented at the

292

IDENTITY AS INGENUITY

Conversation Symposium, Linguistic Society of America Summer Institute, Albuquerque, New Mexico. Goodwin, Marjorie Harness (1990). He-said-she-said: Talk as social organization among black children. Bloomington: Indiana University Press. Hirschberg, Julia, & Diane Litman (1987). Now let's talk about "now": Identifying cue phrases internationally. Paper presented at the annual meeting of the Association for Computational Linguistics, Stanford, California. Jefferson, Gail (1980). The abominable "ne?": A working paper exploration of the post-response pursuit of response. (Occasional Paper No. 6). University of Manchester, Department of Sociology. Levinson, Stephen (1983). Pragmatics. Cambridge: Cambridge University Press. Licari, Carmen, & Stefania Stame (1992). The Italian morphemes no and niente as conversational markers. Acta Linguistica Hungarica 38(1-4): 163-173. Maltz, Daniel N., & Ruth A. Borker (1982). A cultural approach to male-female miscommunication. In John J. Gumperz (ed.), Language and social identity. Cambridge: Cambridge University Press, 196-216. Modan, Gabriella (1994). Pulling apart is coming together: The use and meaning of opposition in the discourse of Jewish American women. In Mary Bueholtz, A. C. Liang, Laurel Sutton, & Caitlin Hines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 501-508. Oliveira e Silva, Giselle M. de, & A. Tavares de Maccdo (1992). Discourse markers in the spoken Portuguese of Rio de Janeiro. Language Variation and Change 4:235-249. Pomerantz, Anita (1984). Agreeing and disagreeing with assessments: Some features of preferred/ dispreferred turn shapes. In J. Maxwell Atkinson & John Heritage (eds.), Structures of social action: Studies in conversation analysis. Cambridge: Cambridge University Press, 57101. Powell, Mava Jo (1992). The systematic development of correlated interpersonal and metalinguistic meanings of stance adverbs. Cognitive Linguistics 3(1):75-100. Sacks, Harvey, Emanuel Schegloff, & Gail Jefferson (1974). A simplest systematics for the organization of turn-taking for conversation. Language 50:696-735. Schegloff, Emanuel, Gail Jefferson, & Harvey Sacks (1977). The preference for self-correction in the organization of repair in conversation. Language 53:361-382. Schegloff, Emanuel, & Harvey Sacks (1973). Opening up closings. Semiotica 8:289-327. Schiffrin, Deborah (1984). Jewish argument as sociability. Language in Society 13:311-335. (1987). Discourse markers. Cambridge: Cambridge University Press. Schwenter, Scott (1996). Some reflections on O SEA: A discourse marker in Spanish. Journal of Pragmatics 25(6):855-874. Thome, Barrie (1993). Gender play: Girls and boys in school. New Brunswick, NJ: Rutgers University Press. Traugott, Elizabeth (1995). The development of discourse markers in English. Paper presented at the Discourse Markers Workshop, Stanford University. Yaeger-Dror, Malcah (1986). Intonational prominence on negatives in English. Language and Speech 28(3): 197-230.

15

A. C. LIANG

Conversationally Implicating Lesbian and Gay Identity

An anecdote While S and her parents were sitting around the kitchen table, her mother reported that in a recent telephone conversation, S's aunt had asked when S was going to get married. S's mother said she had explained that S just hadn't found the right person yet, to which her aunt had replied that not being married was easier anyway. In response to this story, S's father remarked that if S found the right person, she didn't have to rule out marriage. This conversation continued despite the fact that S, who, unbeknownst to her parents, is lesbian, said nothing to indicate her agreement or disagreement. In the absence of S's protestations, S's parents could retain their assumptions regarding how a woman is defined in American culture. Had S elected to express her real feelings—for instance, if she had announced her intention to live with a woman or said that her definition of marriage was quite different from theirs— she might have, minimally, aroused her parents' embarrassment, or she might even have been thrown out of the house. But she chose not to speak, and they could define her sexual identity without feeling contradicted.1 The question of which selves are present in interaction depends on at least three overlapping elements: (1) what constitutes a culturally approved self; (2) the degree to which that self is acceptable to participants, including the speaker and the addressee; and (3) the relative power of each participant within and outside the interaction.2 Culturally approved selves are those that make up a culture's inventory of acceptable identities and roles. For S's parents and aunt, as well as for most people in American culture, and probably the world, part of being a legitimate woman involves marriage to a man; lesbian relationships do not enter the picture. Contrastively, for S (and other lesbians), along with a growing number of heterosexuals, the notion of 'woman' includes acting on emotional and erotic attractions to women, establishing 293

294

IDENTITY AS INGENUITY

committed relationships with them, and so on. This expanded concept of 'woman' opens up the possibility of other nontraditional relationships and identities. In interaction, culturally approved identities are revealed and displayed in accordance with culturally and individually determined expectations of appropriate self-presentations. In addition to global constraints, participants' expectations at the local, interactional level also exert limitations. To the extent that S anticipated a skirmish with her parents and consciously decided not to dispute their assumptions, she shared their comprehension of womanhood as both an insider and an outsider. So strong are normative expectations for a woman in American society that unless an individual indicates otherwise, her identity will be a matter for others, whether in interaction or society as a whole, to settle.3 S's decision not to contest these expectations shows that self-presentation does not occur in a vacuum but is subject to the audience and their expectations and assumptions in relation to the speaker's own. Finally, identities may be imposed by fiat. In this culture, parents may "define" or label their children until their level of autonomy makes such definition inappropriate and unwholesome.4 Institutions may also impose identities on individuals: for instance, the medical system has the power to define individuals in terms of conceptions of illness (as homosexuality was viewed prior to 1975). And as a member of a culture, the individual herself works out a self-definition partly in terms of her understanding of culturally approved identities.

Grice's conversational logic The Western cultural understanding of communication is to be as informative as necessary because knowledge is seen as beneficial (Sweetser 1987). In view of this understanding, H. P. Grice observed that when people are engaged in conversation, they often deviate from informational language use by failing to say what they mean, and yet they remain understandable. Grice (1975) attempted to supply an explanatory theory for how hearers arrive at the meanings conveyed by speakers apart from and beyond the literal meanings of their utterances. He termed his theoretical rubric conversational logic. The mainstay of conversational logic (and indeed of any communicative behavior) is the Cooperative Principle (hereafter abbreviated CP). CP holds that people will make whatever efforts are necessary to calculate another person's intended meanings and to communicate their own meanings so that the hearer understands them. In Grice's words: "Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged" (1975:45). One instantiation of CP consists of the Maxims of Conversation, precepts of conversational behavior observed by interlocutors in accordance with their cultural understanding of language behavior as informative. The maxims include, briefly, injunctions against lying or exaggerating (the maxim of quality), against verbosity or indirectness (the maxim of manner), against offering superfluous or insufficient detail (the maxim of quantity), and against straying from the topic at hand (the maxim of relevance).

Conversationally Implicating Lesbian and Gay Identity

295

Usually, however, conversational contributions are indirect, often superfluous, not entirely truthful, and irrelevant. Expecting that such utterances comply with the maxims at some unspoken level, listeners will try to fill in the gap between the literal and intended meaning of an utterance. Grice termed the process of linking apparently uninformative utterances to their informative equivalents conversational implicature. Since conversational implicature is another way of conveying and interpreting the speaker's meaning, it is also a manifestation of CP. Gender: A basic implicature The presentation of self is one configuration of meanings among many others— including propositions, presuppositions, and entailments; illocutionary and perlocutionary forces; frames; and alignments. Yet among all these meanings, it is unique. In face-to-face interactions, information about gender (along with other social categories such as ethnicity), rather than being stated explicitly, is continually constructed through nonverbal (e.g., gestures, clothing), paralinguistic (e.g., vocal fundamental frequency, intonation), and linguistic signals (e.g., indirectness). A basic cultural implicature is therefore "I am a woman" or "I am a man," which itself evokes a whole set of presuppositions.5 That there is no name for asserting that one is a woman or a man—as the term coming out exists for the assertion that one is lesbian or gay— implies that behavioral expectations associated with womanhood and manhood are normative and pervasive. Thus gender identity appears self-evident. In American and many other cultures, gender is dichotomous: Members of one gender are expected to be attracted to and form coupled relationships with members of the other gender. These assumptions together constitute a fundamental presupposition about gender in this culture, namely, the heterosexual presumption. Gender identity is socially required information for interaction, as relative status among interlocutors is required among speakers of languages such as Japanese that grammatically encode such information. The discomfort aroused when the gender of one's interlocutor is indeterminable or when gender roles are transgressed reveals these categories as a sociocultural organizing principle.6 Because gender is a category that is fundamental to people's beliefs about the world, anyone who questions the foundation of those beliefs is, in some sense, questioning her personal existence. My point here is not to embark on a discussion of the expression of gender per se—there are many excellent works that explore that theme7—but to note that because gender is often implied rather than stated explicitly, it can be categorized in Gricean pragmatic terms as implicature. And because the interpretation of implicature hinges crucially on shared culture-specific presuppositions (Matsumoto 1989; Ochs Keenan 1975), it is useful for the investigation of gender identity. The problem of lesbian and gay self-presentation Homosexuality has thus far not achieved the status of a culturally approved plotline (Polkinghornel 987).8 Consequently, lesbians and gays continually confront the issue

296

IDENTITY AS INGENUITY

of disclosing or concealing their lesbian and gay selves. In every interaction, they must decide whether to reveal their identity and then which strategies they will use to disclose or conceal it. Typically, the situations in which lesbians and gays find themselves fall into three broad categories, with different responses associated with each. In one scenario, self-disclosure would terminate the relationship between the speaker and addressee or, in the worst case, would endanger the speaker's wellbeing. The possibility of disclosure is overridden by considerations of survival (of the relationship or of the individual). Another scenario is one in which the addressee is clearly supportive of or indifferent to the speaker's identity as lesbian or gay. In these contexts, the speaker may self-disclose without compunction. The present concern, however, revolves around the third possibility, in which the speaker is unclear as to the addressee's opinions regarding homosexuality. In this scenario, the speaker is required to consider factors such as her relationship with the addressee and her own socially and culturally shaped beliefs of what is or is not appropriate to disclose. Within American culture, where members value personal truth and sincerity, lesbians and gays are in a double bind. If they observe CP, then they are communicating in good faith because they are making their intentions known. But they are also revealing that they violate gender norms, in which case they risk ostracism, violence or abuse, and discrimination. However, if they fail to observe CP, and, for instance, lie about their sexuality, then they engage in questionable communicative behavior. They may feel guilty for lying or dissembling. Additionally, there is the potential for social sanctions if their nonobservance of CP is discovered later. Cultural norms force on some lesbians and gays a decision that heterosexuals do not have to contend with.

Managing lesbian and gay identity A number of strategies have presented themselves in view of the dilemma faced by lesbians and gays. Erving Goffman offers three possible courses of action in this kind of "normative predicament" faced by members of stigmatized groups: One solution was for a category of persons to support a norm but be defined by themselves and others as not the relevant category to realize the norm and personally to put it into practice. A second solution was for the individual who cannot maintain an identity norm to alienate himself from the community which upholds the norm, or refrain from developing an attachment to the community in the first place. The processes detailed here constitute together a third main solution to the problem of unsustained norms. Through these processes the common ground of norms can be sustained far beyond the circle of those who fully realize them. . . . Passing and covering are involved, providing the student with a special application of the arts of impression management, the arts, basic in social life, through which the individual exerts strategic control over the image of himself and his products that others glean from him. (1963:129-130)

The first solution Goffman proposes may apply, for instance, to lesbians and gays who do not want to be lesbian or gay but who cannot manage to live a normal

Conversationally Implicating Lesbian and Gay Identity

297

heterosexual lifestyle. CP is not in question because such individuals attempt to conform to the norm or at least express commitment to it by repudiating offending behaviors in themselves, as well as in others. The second solution is a form of opting out and includes such strategies as minimizing contact with others to reduce the risk of exposure and abstaining from asking about others' personal lives so as not to invite queries about their own. The third solution that Goffman describes, "passing or covering," may involve lying and, therefore, flagrant violation of CP. It also covers avoidance strategies, which are also violations of CP insofar as the speaker intends to withhold information from the hearer. Because avoidance strategies may also implicate the speaker's sexuality, they can be viewed as cooperative. Such strategies may be less problematic than others, such as outright lies, in terms of both the speaker's own sense of integrity and her need for self-protection. These strategies, in which the status of CP is ambiguous, will be examined next. When CP is ambiguous Ann Weiser (1975) identifies a set of strategies whereby cooperativeness is defined less by the speaker's informativeness than by the degree to which the speaker's intentions can be inferred. She proposes a pair of communicative strategies: conversational devices and conversational stratagems. Both are linguistic means of achieving one's interactional purpose. The former involve direct speech acts, indirect speech acts, and implicature, all of which entail the listener's recognition of the speaker's intention. Use of these strategies signifies the speaker's on-record expressed intention to achieve a given purpose; hence adherence to CP is operative. The latter involve strategies by which the speaker masks her intention to achieve a given purpose. She is not on record as having expressed her intention, and consequently the speaker covertly opts out of CP. For Weiser, then, CP involves not only an exchange of information but also a displayed orientation toward participants' shared goals, whereas nonobservance of CP involves a covert nonorientation toward those goals and this nonorientation appears to promote shared interests. Conversational devices are exemplified by the following dialogue between Jill, a straight woman, and Jack, a gay man: (1) Jill: Jack [who is gay]:

Jack, do you like guys? a. 0 b. Nice weather we're having. c. None of your business. d. What kinda question is that?

In (1), although all of the listed responses differ in degree of directness (responses a and b being the most indirect since they do not explicitly acknowledge the question and hence are most evasive), all are oriented to the question and all conversationally implicate or explicitly express the speaker's hesitance or refusal to answer the question. Silence (a) conversationally implicates reluctance to respond, as does a change

298

IDENTITY AS INGENUITY

of topic (b). Response (c) is an explicit refusal. And the metacommunicative remark in (d) brings the interaction to another level, but in doing so, it communicates the speaker's refusal to answer the question. Here, cooperation involves acknowledgment or orientation to the addressee's remark, although in Robin Tolmach Lakoff's (1995a) terms, (c) and (d) are in clear violation of politeness principles. A conversational stratagem, on the other hand, may arise when a speaker is asked a question that she does not want to answer but is not willing to acknowledge this fact because any response she might undertake would be socially problematic (for example, communicating refusal to answer would implicate her lesbianism). She may therefore pretend that "a sudden thought may have struck, something demanding comment may have occurred in the environment, or the second speaker may not have heard that the first speaker said anything" (Weiser 1975:651). If the questioner believes the response had nothing to do with the question, she will not understand it as a refusal to answer. Furthermore, if she has posed a question with multiple possible interpretations, the speaker can avoid answering without being heard as having refused to respond by responding to an interpretation that the questioner did not intend: (2) Jill: Jack [who is gay]:

Jack, do you like guys? a. Oh no! T forgot to turn off the stove! b. Wow, did you taste how strong this cappuccino is? c. White rice only. I hate brown rice. d. Some of my best friends are guys.

The respondent employs, respectively, the "sudden-thought" device in (a), the "something-presently-demanding-comment" device in (b), the "mishearing" device in (c), and in (d) the "selection-by-reply" device (that is, responding to an unintended interpretation) (Weiser 1975:651). Another way that a conversational stratagem may be achieved is via the construction of an image within which characteristics that might otherwise betray the individual's homosexuality may be reinterpreted as idiosyncrasies. In example (3), Jill, an unmarried woman in her late 30s, may construct an image that emphasizes experiences such as a childhood spent in many different countries, characteristics such as an unusual dialect, or abilities such as ambidexterity. Deviations from the expected life course for a woman would then be seen as another peculiarity about her and not an indicator of her homosexuality. (3) Jack [a naive heterosexual]: Jill [a lesbian]:

Jill, why aren't you married yet? I want to live alone.

Jill's response is interpreted within the context of her overall personality rather than in terms of her homosexuality. Her response is a stratagem because she is encouraging Jack to make the default assumption of heterosexuality. The status of CP differs according to whether conversational devices or conversational stratagems are used. With conversational devices, one or more of the maxims

Conversationally Implicating Lesbian and Gay Identity

299

may be violated, but because CP in some form is upheld, the speaker's meaning corresponds to what is conversationally implicated. In example (1), Jack observes CP to the extent that Jill can calculate an informative meaning even though Jack's response does not conform with the maxims. With conversational stratagems, the listener's ability to calculate the speaker's purpose is suppressed, as in example (2); thus the interlocutors do not share a "set of common purposes or at least a mutually accepted direction." However, the distinction between stratagems and devices is not clear-cut. The two strategies can be functionally identical such that stratagems too can be recognized by listeners as evasive. So the responses in example (2) may also generate implicatures. Consider the following: Joel runs a weekly church discussion group that focuses on minority issues. The group is small, he says, and the conversation is often intimate. Yet Joel has never revealed his own sexuality to anyone in it. "If I were asked directly, 1 might just say, 'Well, we're really not talking about our own orientations here.' Another possible response might be, 'I'm black, I'm a woman, I'm a Muslim, I'm gay, I'm very poor, I am all those things that are discriminated against.' " (Woods & Lucas 1993:143-144)

The first response given by Joel is both a conversational device and a conversational stratagem. It is an overt refusal to answer the question and is legitimate to the extent that the discussion is not focused on him personally. But it is also a way of deflecting attention from himself and so is evasive. The engendering of similar effects from the use of different strategies, known as pragmatic synonymy (Lakoff & Tannen 1979), consequently also renders adherence to CP ambiguous. The following excerpt shows how conversational stratagems may fail and end up being interpreted, correctly, as a desire not to respond. Dave remembers a typical conversation with Audrey, a woman from personnel. "She had a friend who was gay, very blatantly and openly gay, and I know him. One day Audrey came over and says, 'Oh, I didn't know you knew Victor.' Then she says, "How do you know Victor'?' Luckily Victor and I lived in the same apartment building at the time, so I said, 'We live in the same apartment building and there are social functions; that's how I met him.' " The initial dodge seemed to satisfy Audrey's curiosity, but before long she raised the issue again, this time with a question about Dave's roommate (and lover) Kyle. "I guess Audrey put more and more together," Dave says. "I don't know how she found out, but last fall we were walking through the Reading Terminal Market and she asked if I was going to my parents' house or to Kyle's parents' house for Thanksgiving. And I said, 'Well, my parents invited Kyle, but we're going to his parents' house.' " As the conversation continued, Dave grew uncomfortable. Finally, when Audrey asked how long he and Kyle had "been together," Dave responded with an explicit dodge. "She started talking about how long she'd been with her boyfriend, so I finally said, 'Audrey, I'm not going to discuss relationships with you.' I just changed the subject." (Woods & Lucas 1993:142-143)

Whether the stratagem is successfully carried out depends on the hearer's willingness to recognize an implicature or to follow up on the point that was redirected

300

IDENTITY AS INGENUITY

when the conversational stratagem was interjected into the interaction. In the passage above, Dave initially uses a conversational stratagem, the "selection-by-reply" strategy, to respond to Audrey's query. He knows she is asking if he knows Victor because both he and Victor are gay. But because he and Victor both live in the same apartment building, he can exploit this fact and form a response to her query that does not reveal his sexual orientation. However, in a subsequent interaction, she initiates a line of questioning about Dave's relationship with his lover Kyle, which suggests that she was not persuaded by his original strategy. In response, Dave resorts to a conversational device in which he places himself on record as refusing to respond to her question, by which he may have implicated his sexuality. Thus the effectiveness of a conversational stratagem may be considered as much a construction of the hearer as it is of the speaker. (See Rusty Barrett [chapter 16, this volume] for how ethnicity, performed by out-group members, can succeed as a coconstruction of the audience based on the latter's stereotypes of the ethnic group in question.) The tension between having to tell the truth about oneself and having to protect oneself puts some lesbians and gays in a communicatively awkward position. On the one hand, if they employ the strategy of conversational devices, they implicate their sexuality, which involves some degree of risk to their relationships or to their physical well-being (in the case of violence) or their mental well-being (in the case of verbal abuse). If, on the other hand, they successfully employ the strategy of conversational stratagems, they achieve their interactional purposes, such as avoiding having to answer an incriminating question, without conversationally implicating those purposes and without appearing to have opted out of CP. And yet, even in these instances, lesbians and gays often feel deceitful. A fourth solution to the "normative predicament" faced by lesbians and gays is gay implicature. Gay implicature A relationship is cooperative if mutually shared assumptions allow the speaker to conversationally implicate her meaning and the listener to infer it. Cooperative strategies range from close observation of the maxims to the generation of implicature through clear violation of the maxims to overt defiance of the maxims for reasons such as those enumerated in Lakoff (1995a)—to conform with politeness, aesthetics, and so on. Uncooperativeness comprises behavior that adheres neither to politeness principles (whether in observance of the maxims or not) nor to clarity-based communication (insofar as the hearer's ability to calculate the speaker's meaning is impeded). Weiser's (1975) conversational stratagems occupy a possible middle area, for the speaker only gives the impression of being cooperative, although, as mentioned, the conversational stratagem can also be coconstructed by speaker and hearer. Here the avoidance of a delicate subject preserves social relations for both selfprotection (that is, self-defense) and other-protection (that is, politeness), though at the expense of cooperation at the level of the maxims. Given that talk always has, to use William Labov and David Fanshel's (1977) term, immanent reference—that is, that ultimately we are always talking about ourselves—and given that interaction depends on mutual knowledge of the other's so-

Conversationally Implicating Lesbian and Gay Identity

301

cial status (Linde 1993; Matsumoto 1989), the lesbian (or gay) individual is culturally compelled to reveal her (or his) homosexuality or at least finds circumvention difficult. She is also constrained by cultural values or, more precisely, her perception of them as a member of the culture in what she can reveal about herself just in case her interlocutor is hostile to gays. Hence, she must invent conversational strategies that neither implicate her homosexuality, in order to protect herself, nor negate it, in order to conform to the requirement of social interaction that the information she provides about herself adhere more or less to the truth. Confronted with this dilemma of either risking possibly dangerous consequences by coming out or feeling hypocritical in having to lie, lesbians and gays have devised ways of communicating to circumvent it.9 Although these ways of communicating seem to involve deceptive behaviors, often the information from which to draw the correct inferences is there for those who can bring the correct assumptions to bear in an interaction. These communicative strategies comprise what I call gay implicatures. Their covert meanings—though misleadingly worded for "credulous," that is, straight, listeners—can be inferred only if listeners disabuse themselves of the default assumption of heterosexuality. These strategies are unlike Weiserfs conversational stratagems in that listeners can and are expected to infer the speaker's sexuality from what she has conversationally implicated. This expectation is justified, for instance, by the fact that the speaker and hearer live in a place such as the San Francisco Bay Area, where lesbians and gays are publicly visible.10 Thus, to the extent that lesbian and gay speakers implicate their sexuality, they adhere to CP. But because sucessful inference depends on being able to adduce the relevant assumptions, lesbians and gays also circumvent CP when listeners fail in this task. Consequently, just as speaker meanings are coconstructed over the course of interaction, so too is CP. Examples The following are examples of what a hearer needs to know in order to realize the presence of a gay implicature. Genderless reference terms

The first and most common strategy is the use of genderless reference terms such as they or spouse in discussing intimate relationships:11 (4) Scenario: (a) Lesbian: (b) Lesbian:

A lesbian speaker is talking with a (presumably) naive heterosexual. Well the last person 1 was involved with was , they . Well the last person I was involved with was .

This strategy generates a gay implicature because it supplies less information than necessary: Rather than they, why not say he! Because the speaker did not say he, the listener should draw the appropriate inference. In response (b), the speaker uses the relative clause to avoid the gendered pronoun, which should also trigger an im-

302

IDENTITY AS INGENUITY

plicature. The strength of the implicature in both responses (a) and (b) increases when the avoidance of the gendered pronoun persists over the course of several turns. A related strategy for avoiding gendered pronouns is the repeated use of the person. (4') Scenario:

Same as (4).

Lesbian:

I went out on a date with this person, and the person and I got along really well. I'm seeing them again next weekend, etc.

The strength of the heterosexual presumption (along with the speaker's failure to correct wrong assumptions of the gender of the individual being referred to) offsets the lack of cohesiveness entailed by repetition of the noun phrase the person (Halliday & Hasan 1977; cf. Livia, chapter 17, this volume). Once again, the audience's role in the coconstruction of identity is evident. Slightly trickier, however, is the avoidance of gendered terms, even while the listener employs the wrong gendered pronoun. The following dialogue, reported by a lesbian narrator, A, takes place after the narrator's coworker, B, has described her plans to go dancing with a man she has been dating. (5) Scenario'. A: B: A: B: A: B: A:

A lesbian speaker (A) is conversing with her naive heterosexual female co-worker (B), to whom she has not disclosed her sexuality.

I'm looking forward to the weekend. You doing anything special? Well, I'm having a visitor. Ooh . . . that kind of visitor? Does he come in often? Actually, yes. . . . Is this someone special? I think so ... we'll see.

To draw the implicature, the listener must note that the speaker avoids referring to the visitor with gendered pronouns. Thus the omission of expected information may also trigger gay implicature. However, Grice never discusses the issue of whether allowing the listener to draw the wrong implicature is an instance of uncooperative behavior. So although the lesbian speaker herself does not state anything to commit to the heterosexual presumption, and therefore may be said to be conveying a gay implicature, one can imagine B, on discovery of her lesbian co-worker's homosexuality, saying indignantly, "Why did you let me go on thinking that?" The possibility that B may feel deceived suggests that the lesbian speaker may not be observing CP. "Not my type" Another strategy involves narrowly construing queries concerning romantic interests, as in (6):

Conversationally Implicating Lesbian and Gay Identity (6)

Scenario:

B: L: B: L: B: L:

303

B, a naive heterosexual, is inquiring of a lesbian speaker, L, about her availability.

So, dating any men? No. Why not? I'm not interested in finding a man. Just haven't found a good enough man, eh? No, I'm just not interested in finding a man.

In this example, the statement I'm not interested in finding a man should be understood as conversationally implicating interest either in finding a woman or in finding no one at all. Stress can signal a difference in the meaning that is implicated. Thus putting the stress on find, as in I'm not interested in finding a man, leaves the heterosexual presumption intact by rejecting the proposition in its entirety. If the emphasis is placed on man, as in I'm not interested in finding a man, contrastive stress signals that the speaker is interested in finding something else, namely a woman. Gay implicature appears to be generated when the stress is placed on interested, as in I'm not interested in finding a man, because it leaves the crucial part of the proposition, finding a man, neutral. Other strategies (7)

Scenario:

B: L:

A naive heterosexual male (B), who is potentially interested in the lesbian speaker, queries her about her availability. The lesbian speaker (L) has been dating another lesbian, but she also has conflicts about being lesbian, that is, she would rather be heterosexual.

Have you been seeing anyone recently? (a) I haven't been seeing any men recently, (b) I haven't been seeing anyone recently.

Finally, in (7), a lesbian was asked by a prospective male suitor, B, "Have you been seeing anyone lately?" Although she uses a strategy categorizable as gay implicature, her self-presentation is ambiguous. She could have lied and stated that she had not been seeing anyone romantically. More subtly, she might have responded with either an elided form, No, I haven't, or a repetition of the proposition, No, I haven't been seeing anyone lately, and be understood as having said she had not been seeing anyone of the other sex, without implicating anything about the same sex. However, such a response presupposes the default assumption of heterosexuality. Because she could not honestly maintain that assumption, she could not give either negative response without opting out, unless she explicitly indicated that a maxim had just been violated. In this sense, she did not lie. Yet neither did she tell the truth since she could have replied with a version of an affirmative response. Hence, she replied, "I haven't been seeing any men recently" with no contrastive stress placed on any men. By stating the obvious, that is, by violating the maxim of manner, she answered the suitor's

304

IDENTITY AS INGENUITY

question without compromising her moral stance. Her statement that she had not been seeing any men lately should have triggered the less obvious unstated alternative, namely that she had been seeing a woman. Therefore, the listener should have concluded that she was lesbian. "So," she said, pursuing her reasoning to its conversationally logical conclusion, "I came out to him." He seemed satisfied with her reply, as evidenced by ensuing dates, while the speaker had maintained her commitment to cooperativeness, making what she termed her self-disclosure apparent (to anyone who listened closely). Note that if she had given response (b), there would likely be no implicature. The reason for the difference is that response (b) acquires its function as gay implicature in relation to what was said previously. Contrasting with the suitor's use of anyone, the lesbian's use of any men in a sense defeats the implicature of any men in the former's use of anyone and triggers another, that is, gay, implicature. In response (b), however, there is no implicature generated by the suitor's any men, and hence, none in the repetition of the corresponding noun phrase by the lesbian. Example (7) attests to how complex identity is, and how ambiguity itself can be polysemous. In examples (4) through (6), individuals implicated their unwillingness to enact a heterosexual identity, but they did not preclude false conclusions about their sexuality. If listeners maintained the expectation of heteronormativity, the lesbian or gay speaker did nothing overt to disabuse them of their erroneous conclusion. Their identities were in this sense ambiguous. However, in (7), the speaker employs gay implicature as a way of being different things to different people (e.g., arguably straight to her suitor and lesbian to the lesbian interlocutor to whom she reported the dialogue). If she had truly believed that she were or could be heterosexual, she could merely have stated that she had not been seeing anyone lately, which could have been interpreted as fully cooperative and informative. That is, her intention to be heterosexual would render such a statement, which would implicate not having seen someone of the other sex, as completely relevant and as providing just the information required by her interlocutor, even if she were exploiting the heterosexual presumption. She would thereby eliminate any possibility of gay implicature. Yet the fact that she did not do so and, further, the fact that she believed that she "came out" to the suitor suggest that her self-concept is lesbian. Thus her presentation of an ambiguous self was not merely a conscious strategy adopted out of selfdefense or the need to perceive herself as a moral person. The speaker was, in addition, attempting to have it both ways by retaining the option of being either straight (by allowing the suitor to believe that she was available) or lesbian (by nonetheless implicating her homosexuality).

Ambiguity: Lesbian/gay identity and the Cooperative Principle The selves presented by lesbians and gays who engage in the use of gay implicature implies noncommitment to, and thus has the quality of ambiguity in relation to, accepted (heterosexual-based) gender categories. A possible motivation for gay implicature is a model of the self by which the speaker tries, to the extent possible, to

Conversationally Implicating Lesbian and Gay Identity

305

be honest and sincere in her interactions rather than one by which she attempts to do what is necessary to conform, even if it means violating CP. Gay implicature is less socially problematic than other violations of interactional rules. Although words exist for other discursive acts of lesbian and gay self-presentation (e.g., coming out, coming-out stories), no such culturally recognized term exists for gay implicature even though it is a common form of discourse among lesbians and gays. Likewise, there are words for infractions against CP, such as lie, fib, falsehood, and those who commit them, such as liar, perjurer, prevaricator (cf. Verschueren 1979). But it is more difficult to think of words describing conversational behaviors that suppress the addressee's ability to calculate a speaker's intentions or even give the addressee the option of calculating them. The absence of any such terms for gay implicature suggests that the way the culture evalutes such strategies is open and indeed ambiguous. Inasmuch as such strategies are less accessible to naming and therefore to value judgments, the moral implications are less problematic than for maxim violations.12 Thus it cannot be said that gay implicature (and conversational stratagems, which are pragmatically homonymous with utterances that give rise to implicatures) are overtly deceptive; moreover, the relevant information is in the message, and the speaker is acting out of self-defense. To this extent, then, speakers employing these strategies are "passing" by means of their ambiguous adherence to CP. Because the strategies present an ambiguous self, and because the speaker does not seek confirmation for whether or not the audience has detected the speaker's ploy, the possibility exists that the audience is able to infer the speaker's sexuality. Goffman observes that those who are stigmatized and those who are normative may cooperate in the construction of "normativity" of the stigmatized. Also involved is a form of tacit cooperation between normals and the stigmatized: the deviator can afford to remain attached to the norm because others are careful to respect his secret, pass lightly over its disclosure, or disattend evidence which prevents a secret from being made of it; these others, in turn, can afford to extend this tactfulness because the stigmatized will voluntarily refrain from pushing claims for acceptance much past the point normals find comfortable. (Goffman 1963:129-130)

Consequently, what hearers are actually aware of and how much they actually cooperate when the speaker uses such strategies is not necessarily known. What is important is when the speaker's ambiguous self-presentation goes unchallenged. In such instances, it is not unreasonable to speculate that CP may be observed by both parties, whereby norms remain superficially intact even if the deviator is in (tentative) violation of them. Conclusion Identity is not constant but emerges from the interactional flow through which shared meanings are negotiated between the speaker, whose intentions to self-presentation are culturally and individually determined, and the addressee, whose expectations

306

IDENTITY AS INGENUITY

of possible and legitimate identities likewise combine personal and cultural beliefs. Depending on which assumptions are shared, certain implicatures, and hence identities, are more likely to be conveyed and inferred than others. Gay implicature is one site where language, cultural beliefs, and identity intersect. By invoking this conversational strategy, lesbian and gay speakers can exploit addressees' cultural expectations of legitimate persons in order to present an ambiguous identity. Strategies such as conversational stratagems and gay implicature suggest that adherence to CP cannot be defined as the calculability of default implicatures. Adherence to CP, and thus the calculability of implicatures, also depends on other real-life considerations like relative social status, face, power relations, and politeness principles. As noted, conversational stratagems that are ostensibly CP violations can be ambiguous with utterances that give rise to implicatures; in such instances, whether the speaker has violated CP is as ambiguous as the stratagem itself. The status of CP, then, is not simply one of speaker adherence or nonadherence. In regard to gay implicature, the hearer's ability to calculate the speaker's intended meaning is a result of an ongoing and perhaps implicit process of negotiation between the speaker and hearer. Identity results from the interaction between the individual's perceived cultural expectations of what a "normal" human being is and her own subjective experiences in relation to those expectations. Cultural imperatives in the form of shared expectations of what is "natural" place individuals who cannot conform with them (or who conform more easily with negatively evaluated social categories) in psychologically difficult situations because the individual encounters a predicament through which, for example, her subjective experience of same-sex attraction prevents her from realizing the norm of heterosexuality. Thus cultural imperatives impose moral dilemmas that are not faced by "normative" individuals within the culture: Lesbians and gays must "decide" whether to present a self that conforms with or violates cultural expectations. A naive hearer may fail to calculate a gay implicature, and in some formulations of CP (such as Lakoff's), the speaker may be said to be uncooperative. Where the community as a whole has had exposure to lesbians and gays, the speaker's continued use of gay implicature over the course of an interaction may lead the hearer to make the correct inference. In this case, based on the hearer's success in decoding the speaker's intended meaning, the speaker observes CP without changing her intentions regarding the calculability of her homosexuality. Over the course of an interaction, gay implicature can implicate a noncommitment to the heterosexual presumption, which can eventually lead the hearer to infer the speaker's homosexuality.13 Finally, the status of lesbian or gay identity as a culturally acceptable identity can be assessed by the speaker's degree of adherence to the maxims in communicating her sexuality. In the act of coming out, for instance, the maxims are observed and there is no room for the audience to do any inferencing. Coming out presupposes that the audience subscribes to the normative model of gender and sexuality and that the most effective way to dislodge at least one component of this model, the heterosexual presumption, is through clarity-based communication. The distress that such directness causes to those who maintain the heterosexual presumption implies that the two underlying models of gender (that individuals are meant to be coupled with the other sex versus the absence of such a constraint on choice of mate) are unshared by speaker and

Conversationally Implicating Lesbian and Gay Identity

307

audience; this cognitive mismatch is also evidenced in the numerous books that have been published and organizations that have been established whose main purpose is to deal with the consequences of this communicative act. The emergence of gay implicature demonstrates that the status of lesbian and gay identities on a culture-wide level is changing. The mere possibility of decoding utterances that implicate the speaker's sexuality suggests that the heterosexual presumption has been disengaged from current cultural presuppositions regarding gender. If homosexuality were as taboo as, say, pedophilia, the odds would be very slight that the speaker would allow identification of these tendencies. Given the illicit status of pedophilia, he would lie, thereby committing a clear infraction against CP. Gay implicature, in contrast, perhaps indicates lesbians' and gays' hope that the heterosexual presumption is losing ground. Lesbians and gays are in the peculiar position of being at the forefront of cultural change, depending on whether they decide to lie (thereby maintaining the status quo), to tell the truth (thereby forcing, at the risk of physical or psychological self-harm, another person to confront her conception of normality), or to convey their identity through gay implicature (thereby avoiding commitment to the old normative category and initiating a process of "training" in the new nonnormative lesbian or gay category). The flip side of this moral dilemma also offers "choices" between possible self-presentations. If lesbians and gays give the appearance of being heterosexual (by "passing" or even adopting all the external trappings of heterosexuality, including living a wholly heterosexual lifestyle) or of being ambiguously heterosexual and homosexual, in some sense (one exploited by antihomosexual political groups) identity is conscious and therefore voluntary. Yet this reasoning fails to differentiate between performed (social) identities and biographical (personal) identities, a disjunction that lesbians and gays, because of the absence of inherent identifying traits, make salient. The range of self-presentations available to lesbians and gays merely exemplifies the performed nature of all identities and social categories, whether or not they correspond to the individual's biographical self. Lesbians and gays who "pass" as straight perform an identity that is dissonant with who they actually experience themselves to be. Identities, both stated and implicated, are a result of linguistic decisions made over the course of an interaction, in which each individual's contribution reflects both the individual's intention and the local and global social constraints. This process of negotiation between individuals, intentions, and social and cultural forces illustrates the intimate connection between language and identity. NOTES

I am grateful to Robin Tolmach LakofT and Sara Gesuato for their assistance in elucidating the theoretical basis of this chapter. Pamela Morgan, Collin Baker, Mary Bucholtz, Susan Ervin-Tripp, and Michael Meacham have my thanks for providing ample and useful comments. I also wish to thank the unnamed individuals who contributed their stories. Any shortcomings are my own. 1. I thank Christopher Liang for this anecdote. 2. For ease of discussion, I consider the presentation of self from the perspective of the speaker, but the addressee also displays a self through her listening behaviors. Whether speak-

308

IDENTITY AS INGENUITY

ing or silent, one cannot not communicate in social interaction (Bateson 1972); consequently, one cannot not present a self. 3. That identity hinges on power is illustrated by the question of how Jewishness is defined. Strictly speaking, according to Judaic law, only if the mother is Jewish do the offspring "count" as Jewish. Yet even those who do not themselves identify as Jewish recognize that Jewishness is contingent on who has the power to characterize those who are and those who are not. Were a right-wing white supremacist anti-Semitic group to seize control of the country, it is likely that anyone with Jewish blood would be defined as Jewish, regardless of the individual's own self-identifications or of what Judaic law states. 4. Some diseases with psychological components (cf. Bateson 1972 on schizophrenia) are viewed as a result of the ways families impose a role on a family member in order to ensure survival of the larger family unit. See also Lisa Capps (chapter 4, this volume). 5. In Robin Tolmach Lakoff's construal of Grice's conversational logic (the system of precepts or maxims adhered to by speakers and listeners in producing and interpreting utterances in conversation), "maxim-observant utterances do exactly and succinctly express pure semantic meaning; but they may not incorporate many of the pragmatic signals that orient participants to significant aspects of the message: discourse genre, deictic situation, seriousness, level of intimacy, mutuality of trust, delicacy of subject matter, and much more. Implicature provides that information, often as important in the full understanding of a communication as its explicit denotation" (1995b:191). To Lakoff's specification of the meanings covered by implicature, I would add culturally approved, situation-appropriate identities. 6. The discovery that Brandon Teena, a preoperative transgcndered man living in Humboldt, Nebraska, was biologically female aroused the rage of two of his former acquaintances. In 1993, Teena was arrested on misdemeanor charges for check forgery. The information, including the fact that he was anatomically and legally female, was released by police to the local newspaper, the Falls City Journal. A week later, Teena was raped by two men who threatened to kill him if he went to the police. Although he reported the rape to the police and identified the two assailants, no charges were filed against them. Teena was subsequently hunted down by them and murdered, along with his female lover and another friend. 7. See, for example, the chapters by Rusty Barrett, Jennifer Coates, Marjorie Faulstich Orellana, Deborah Tannen, and Keith Walters in this volume and numerous contributions in Anna Livia and Kira Hall (1997). 8. Cf. also Mandy Aftel (1996), William Leap (chapter 13, this volume), Charlotte Linde (1993), Theodore Sarbin (1987), and Kathleen Wood (chapter 2, this volume) for other narrative-based conceptions of culture-specific identities. 9. Perhaps another factor compelling the use of gay implicature rather than bald-faced lies is lesbians' and gays' resentment of the double standard: When lesbians and gays merely state their sexual orientation, they are often viewed by heterosexuals as "flaunting" their lifestyles. Susan Ervin-Tripp (personal communication) has recounted the story of a heterosexual colleague who objected to a pink triangle sticker on the door of the office of a fellow academic who was lesbian. According to the colleague, the lesbian academic's sexuality was irrelevant to her capabilities as a researcher and teacher. (Presumably, this same colleague did not consider that wearing a wedding band was equally irrelevant to the qualifications of an academic.) Gay implicature enables lesbians and gays to thwart the proscription against "flaunting" it by refusing to commit to the heterosexual presumption, albeit covertly. 10. However, it is also the case that some naive heterosexuals will never get the implicature, no matter how many times it is repeated, whether due to willful blindness, lack of experience, or other factors. 11. Examples (4) through (7) were provided by members of several lesbian and gay electronic-mail lists. From a methodological and theoretical point of view, these data may be

Conversationally Implicating Lesbian and Gay Identity

309

problematic insofar as they may neither reflect what actually happened—that is, as selfreports, the examples are tantamount to being constructed examples—nor be representative of the behavior of lesbians and gays as a group. Nonetheless, the examples cited do indicate the extent to which the devices in question are consciously aimed toward the avoidance and transformation of a difficult social situation. 12. Regardless, behavior that is less accessible to naming and therefore to a value judgment is not necessarily unproblematic for the hearer. It is in fact convenient for the speaker that there are no ready-made names or value judgments to condemn the suppression of the hearer's ability to calculate the speaker's intention. 13. Sara Gesuato (personal communication) has given me the analogous example of the use of signora versus signorina in Italy. In recent years, it has become a law that all women over eighteen years of age can adopt the title signora whether or not they are in fact married. The title has become the Italian equivalent of the American English Ms. Signorina meanwhile retains the meaning of Miss. According to Gesuato, since the law has been effective for several years, her use of the title signora over signorina should convey to strangers that she is not willing to state her marital status (and that, furthermore, they should not inquire). The fact that some of her fellow Italians continue to ask about her marital status indicates that the law and its social significance for women have only partly been incorporated into Italians' collective consciousness. REFERENCES

Aftel, Mandy (1996). The story of your life: Becoming the author of your experience. New York: Simon & Schuster. Bateson, Gregory (1972). Steps to an ecology of mind. New York: Ballantine. Goffman, Erving (1963). Stigma: Notes on the management of spoiled identity. Englewood Cliffs, NJ: Prentice Hall. Grice, H. P. (1975). Logic and conversation. In Peter Cole & Jerry L. Morgan (eds.), Syntax and semantics. Vol. 3: Speech acts. New York: Academic Press, 41-58. Halliday, Michael A. K., & Ruqaiya Hasan (1977). Cohesion in English. London: Longman. Labov, William, & David Fanshel (1977). Therapeutic discourse. New York: Academic Press. Lakoff, Robin Tolmach (1995a). Conversational implicature. In Jef Verschueren, Jan-Ola Ostman, & Jan Blommaert (eds.), Handbook oj pragmatics. Amsterdam: John Benjamins, 1—11. (1995b). Conversational logic. In Jef Verschueren, Jan-Ola Ostman, & Jan Blommaert (eds.), Handbook of pragmatics. Amsterdam: John Benjamins, 190-198. Lakoff, Robin Tolmach, & Deborah Tannen (1979). Communicative strategies in conversation: The case of Scenes from a Marriage. In Christine Chiarello et al. (eds.), Proceedings of the Tenth Meeting of the Berkeley Linguistics Society. Berkeley, CA: Berkeley Linguistics Society, 581-592. Liang, A. C. (1995). Gay implicature as straight delusion. Paper presented at the annual meeting of the American Association for Applied Linguistics, Long Beach, California. Linde, Charlotte (1993). Life stories: The creation of coherence. Oxford: Oxford University Press. Livia, Anna, & Kira Hall (eds.) (1997). Queerly phrased: Language, gender, and sexuality. New York: Oxford University Press. Matsumoto, Yoshiko (1989). Reexamination of the universality of face: Politeness phenomena in Japanese. Journal of Pragmatics 14:237—261. Ochs Keenan, Elinor (1975). The universality of conversational implicature. In Ralph Fasold & Roger Shuy (eds.), Studies in language variation. Washington, DC: Georgetown University Press, 255-268. Polkinghorne, Donald E. (1987). Narrative knowing and the human sciences. Albany: SUNY Press.

310

IDENTITY AS INGENUITY

Sarbin, Theodore (ed.) (1987). Narrative psychology: The storied nature of human conduct. Chicago: Praeger. Sweetser, Eve (1987). The definition of lie. In Dorothy Holland & Naomi Quinn (eds.), Cultural models in language and thought. Cambridge: Cambridge University Press, 101-127. Verschueren, Jef (1979). What people say they do with words. Ph.D. diss., University of California, Berkeley. Weiser, Ann (1975). How to not answer a question: Purposive devices in conversational strategy. In Robin E. Grossman, L. James San, & Timothy J. Vance (eds.), Papers from the Eleventh Regional Meeting of the Chicago Linguistic Society. Chicago: Chicago Linguistic Society, 649-660. Woods, James D., & Jay H. Lucas (1993). The corporate closet: The professional lives of gay men in America. New York: Free Press.

Part IV IDENTITY AS IMPROVISATION

This page intentionally left blank

16

RUSTY BARRETT

Indexing Polyphonous Identity in the Speech of African American Drag Queens

n this chapter, I examine the presence and use of a "white-woman" style of speaking among African American drag queens (hereafter AADQs). I hope to demonstrate that a close examination of this language use suggests an ambivalent, sometimes critical, sometimes angry, view of whiteness that does not lend itself to a simplistic explanation of "wanting to be white." After discussing the issue of drag itself, I will discuss the ways in which AADQs create a "white-woman" linguistic style. However, this style of speaking is only one voice used by AADQs. The complete set of linguistic styles together index a multilayered identity that is sometimes strongly political with regard to issues of racism and homophobia.

Drag Before discussing the issue of drag, it is important to distinguish drag queens from other transgender groups, such as transsexuals, transvestites, cross-dressers, and female impersonators.1 Transsexuals are individuals who feel that their gender identity does not correspond to the sex that they were assigned at birth. Many (but not all) transsexuals undergo hormone treatment or "sex-reassignment" surgery as means of altering their physical appearance to match that typically associated with their gender identity. Transsexuality and homosexuality are independent issues, and transsexuals may be heterosexual, homosexual, bisexual, or asexual (cf. MacKenzie 1994). In contrast, drag queens do not identify themselves as having female gender (that is, they do not see themselves as women). 313

31 4

IDENTITY AS IMPROVISATION

Unlike transsexuals, transvestites identify with the gender corresponding to their assigned sex. The terms transvestite and cross-dresser refer to anyone who wears clothing associated with the other gender. The term transvestite does not necessarily refer to an individual who fully crosses gender roles, and it may be used for situations such as a man wearing women's undergarments under traditional "male" clothing. Studies have suggested that between 72 and 97 percent of male transvestites are heterosexuals (Bullough & Bullough 1993). In contrast, drag queen refers almost exclusively to gay men (with drag king referring to lesbian cross-dressers). The term female impersonator is very similar to drag queen, although (like transvestite and cross-dresser) it may be used to refer to heterosexuals. Female impersonators are professional cross-dressers who typically focus their performances on creating a highly realistic likeness of a famous woman (such as Diana Ross, Cher, Reba McEntire, or Madonna). Glamour-oriented drag queens (or "glam queens") often produce a physical representation of hyperfeminine womanhood that is quite similar to that of female impersonators. The performances of female impersonators generally build on their ability to "pass" as women, however, and drag queens usually make no pretense about the fact that they are (gay) men, even though they may present a realistic image of a particular type of woman (cf. Fleisher 1996:14-15). Also, female impersonators generally perform for the amusement of heterosexuals, whereas drag queens perform for lesbian and gay audiences. Although both female impersonators and drag queens may produce highly similar external conceptions of femininity, the intent and attitude behind their performances are quite different. All the drag queens in this study are glam queens. They typically go to great lengths to produce a highly feminine image. In addition to wigs, makeup, and "tucking" (hiding one's genitals), drag queens often use duct tape to push their pectoral muscles closer together to give the impression of cleavage. Glam queens almost always wear high-heeled shoes and shave their arms, legs, chest, and (if necessary) back. The dresses worn by glam queens are quite extravagant, often covered in beads or sequins. Many dresses do not have sleeves or have high slits to make it clear that the wearer is not trying to hide masculine features under clothing. Jewelry is almost always worn, especially large earrings and bracelets. The overall goal is to produce an image of hyperfemininity that is believable—an image that could "pass" for a woman. The ideal of glam drag is to be "flawless," or to have no visual hints of masculinity that could leave one open to being "read" (insulted; see also Morgan, chapter 1, this volume). The drag queens I studied are professional entertainers who work primarily in gay bars. In order to become a full-time professional, a drag queen must achieve a certain degree of exposure, usually by working without pay or by winning beauty pageants. Thus, to become a professional, a drag queen must prove that she is sufficiently flawless.2 Drag queens who are not flawless may be viewed as "messy": lacking professionalism both in the image produced and in the demeanor presented in the bar. Thus, a messy queen is often one who is unsuccessful at presenting an image of "proper" femininity (both in speech and poise). The term messy may also be used for queens who cause problems by spreading gossip or getting into trouble through drugs, alcohol, theft, or prostitution. A messy queen has little chance for success as

Indexing Polyphonous Identity

31 5

a professional performer because she is unable to convey a convincing image of femininity both on stage and during interactions in the bar. Feminist scholars have argued that drag is inherently a misogynistic act, primarily because they feel that it represents a mockery of women or, at the very least, a highly stereotyped image of femininity and womanhood (Ackroyd 1979; Frye 1983; Lurie 1981; Raymond 1994, 1996; Williamson 1986). It has also been argued that drag is a way of reinforcing a performer's masculinity by demonstrating that he is not actually a woman but that he is able to control the qualities associated with women (Gilbert 1982; Showalter 1983). Because the goal of glam drag is to produce an outward appearance indistinguishable from that of a "real" woman, humor in the performance of glam drag is not derived from the performer's inability to "be" a woman but from the virtuoso performance itself. The argument that drag is primarily a mockery of women relies on the stereotyped perception of drag queens displaying "big tits, fat tummies, wobbly hips and elaborate hair-dos" (Williamson 1986:48) that "draw hoots and howls in audiences of mostly men" (Raymond 1996:217). With the exception of elaborate hairstyles, this stereotyped image of drag has very little to do with the reality of the gay drag performances included in this study. The drag performers I studied do not intend to produce laughter through their appearance. As Edmund White has argued, drag (at least among gays) "is an art of impersonation, not an act of deception, still less of ridicule" (1980:240). These arguments against drag often confuse gay drag queens with the sort of transvestite shows produced by straight (usually white and wealthy) men as a sort of male bonding experience, even though the latter (often including hairy men wearing exaggerated false breasts and rear ends) are quite different in both content and intent. More recently, commentators (Butler 1990, 1993; Feinberg 1996; Fleisher 1996; Hilbert 1995) have critiqued this perspective not only because it views all forms of transgender behavior as male homosexual activities but also because it places women at the center of male homosexuality. These scholars argue that drag is not "about" women but rather about the inversion or subversion of traditional gender roles. These scholars often praise drag queens for demonstrating that gender displays do not necessarily correlate with anatomical sex and typically see drag as a highly subversive act that deconstructs traditional assumptions concerning gender identity. Butler, for example, argues that drag exposes the imitative nature of gender, showing that gender is an "imitation without an origin" (1990:138). Rather than viewing drag as an imitation of women, queer theorists usually glorify it as a highly political deconstructive force working to undermine gender assumptions. Drag queens themselves also adamantly reject the notion that drag mocks women. They distinguish their performances from those of heterosexual men (who, in their view, clearly do mock women). For example, the nationally known AADQ singer RuPaul was angry and offended when she had to copresent an MTV music award with Milton Berle, an older heterosexual comedian known for using drag in his humor. She reports in her autobiography that problems between her and Berle began when he insulted her backstage (RuPaul 1995). The problems continued onstage as well, resulting in an argument that was aired on live television. According to RuPaul,

31 6

IDENTITY AS IMPROVISATION

the producers of the program did not realize that there was no connection between her status as a drag queen and Berle's use of women's clothing to produce humor at women's expense. As she describes it, "They didn't get that my take on drag is all about love, saying that we are all drag queens. It's certainly not about putting women down. And it's not about being the butt of a bunch of cheap dick jokes" (1995:181). In addition, drag queens sometimes see themselves as fighting against gender oppression in general, a cause that many feel should garner support rather than disdain from feminists. And despite the role of drag queens in the gay liberation movement (cf. Duberman 1993; Marcus 1992), many gay men openly express scorn for drag queens. Hapi Phace, a New York drag queen interviewed by Julian Fleisher (1996), points out, "The thing that you have to remember is that as drag queens we have a lot of the same issues as feminists in our own dealings with the gay community. To gay men, we're considered 'women.' We get to see a lot of the misogyny in gay men" (Fleisher 1996:33-34). This view sees drag performers (both kings and queens) as part of a larger set of individuals persecuted by an intolerant society for their deviance from prescribed gender norms. As Leslie Feinberg argues, "it's really only drag performance when it's transgender people who are facing the footlights . . . the essence of drag performance is not impersonation of the opposite sex. It is the cultural presentation of an oppressed gender expression" (1996:115). In other words, drag is not intended as a negative portrayal of "women" but rather is an expression of a particular gender performance (cf. Butler 1990)—a performance by those who are themselves oppressed by the forces of patriarchy. Part of the fascination with drag is its ability to cause such diverse reactions in different contexts and with different audiences. In some instances, cross-dressing is used as a weapon of misogyny and even homophobia. In other contexts, drag may serve to question the rigidity of prescriptive gender roles, acting as a tool of liberation. One of the main functions of drag performance is to expose the disunity between perceived or performed identity and underlying "authentic" biographical identity. The "meaning" of drag is often created by audience members in their individual attempts to reconnect their physical perceptions of the performance with their personal assumptions concerning social identity and gender categories. Many drag queens argue that they are not really trying to "achieve" any great social message but are merely expressing their personal identity (which happens to involve cross-dressing). The celebration and even glorification of drag by queer theorists such as Butler might be seen as exploiting drag-queen identity for the sake of theoretical deconstruction of gender categories. Like the feminist view of drag as inherently misogynistic, the view that drag is inherently subversive imposes a unidimensional meaning on the personal identity of a particular group. But there are certainly cases in which drag-queen performances are clearly misogynistic. As Miss Understood, another of Fleisher's interviewees, argues, "I think that men in general are pretty misogynist. Men are sexist all the time and if drag queens are men, of course there's going to be sexist things coming out of their mouths" (Fleisher 1996:32). Although drag queens may be misogynistic at times, their personal identity as drag queens does not make them de facto sexists. In many cases, they may be viewed as highly subversive. Thus neither the view of drag as inherently subversive nor as inherently misogynistic is

Indexing Polyphonous Identity

317

"correct." Rather, drag queens are individuals whose social identity no more determines their political stance than any other aspect of their personal identities, such as gender, class, or ethnicity. Indeed, the performances by AADQs considered here generally focus on other aspects of identity (such as sexual orientation, ethnicity, and class) rather than on the issue of cross-dressing itself.

Polyphonous identity and acts of performed identity Historically, sociolinguistic studies have tended to view identity monolithically, often assuming a one-to-one relationship between language use and membership in some identity category (usually based on class, race, or sex). Speakers were "allowed" only a single identity that was typically mapped onto a particular identity category. Those who did not fit the norms of language usage were implicitly viewed as possessing a "failed" identity, as with William Labov's (1972) lames or Peter Trudgill's (1983) concept of conflicting identity. Thus the fact that some speakers could not easily be classified into a particular identity category on the basis of their language usage was seen as a problem with the speaker rather than a problem in the research paradigm. As Marcyliena Morgan (chapter 1, this volume) points out, Labov's focus on unemployed adolescent boys in his study of African American Vernacular English (hereafter AAVE) has contributed to stereotypes of what constitutes a "real" African American identity. Sociolinguistic research has typically perpetuated the myth that one must speak AAVE (and must usually be a heterosexual male) to qualify as a "true" African American, leaving many African Americans classified as "lames" or simply ignored. This myth of what constitutes African American identity is especially relevant to African American gay men. Because of the combined forces of racism in the white gay community (cf. Beame 1983; Boykin 1996; DeMarco 1983) and homophobia in the African American community (Boykin 1996; hooks 1989; Monteiro & Fuqua 1994), African American gay men are often pressured to "decide" between identifying with African Americans or with white gay men (Peterson 1992; Simmons 1991; Smith 1986; Tinney 1986). Due to the stereotypical view that AAVE is somehow tied exclusively to young heterosexual men and is a strong marker of masculinity (cf. Harper 1993; Walters 1996), the use of "Standard" English by African American gay men (including drag queens) contributes to the argument that they have somehow abandoned the African American community by identifying themselves as gay. Thus simplistic conceptions of the relationship between language and identity in sociolinguistic research may serve to reinforce the racism and homophobia prevalent in American society. More recently, as studies in language and gender have moved to a practicebased approach (Eckert & McConnell-Ginet 1992), it has become clear that identities based on categories such as gender, class, and ethnicity are often enmeshed in very complex ways. Expressions of gender are simultaneously expressions of ethnicity (Bucholtz 1995; Hall 1995) and of class (Bucholtz, chapter 18, this volume; McElhinny 1995; Woolard 1995). Hence the concept of a prescriptive norm for "women's language" is often a reflection of ideology concerning not only gender but also race and class.

31 8

IDENTITY AS IMPROVISATION

Given the complex relationship between linguistic form and ideologies of gender, class, race, and ethnicity, one would expect speakers to attune their linguistic performances to their personal stance toward gender and other ideologies. Speakers may heighten or diminish linguistic displays that index various aspects of their identities according to the context of an utterance and the specific goals they are trying to achieve. Thus a speaker may use the indexical value of language (cf. Ochs 1992) to "position" (Davies & Harre 1990) the self within a particular identity at a particular interactional moment. This practice implies that speakers do not have a single "identity" but rather something closer to what Paul Kroskrity (1993:206 ff.) has called a "repertoire of identity," in which any of a multiplicity of identities may be fronted at a particular moment. In addition, at any given moment speakers may also convey more than one particular "categorical" identity. For this reason I have chosen the term polyphonous identity rather than repertoire to convey the idea that linguistic displays of identity are often multivoiced or heteroglossic in the sense of Mikhail Bakhtin (1981, .1984). Thus speakers may index a polyphonous, multilayered identity by using linguistic variables with indexical associations to more than one social category. In the case of AADQs, speakers typically use language to index their identities as African Americans, as gay men, and as drag queens. Through style shifting, the linguistic variables associated with each aspect of identity may co-occur, creating a voice simultaneously associated with several identity categories (cf. Barrett 1998). One important distinction between the language of drag-queen performances and many other forms of language is that although drag queens use language to index "female" gender, they do not generally see themselves as "women." Thus they perform an identity (as a "woman") that they may see as distinct and separate from their own biographical identity. Sociolinguistic theory has not traditionally made a distinction between a performed identity and those identities associated with the social categorization of the self. In her analysis of drag, Butler (1990) points out that, in addition to the traditional distinction made between sex and gender, drag creates the need for a third category, performance. Although gender performance often corresponds directly with gender identity, cases such as drag require an understanding that performed gender may differ from self-categorized gender identity. The majority of drag queens maintain "male" gender identity alongside "female" gender performance. Indeed, perhaps the strongest distinction between drag queens and transsexuals is the distinction between performance and identity, in that transsexuals typically maintain a gender identity that corresponds to their gender performance (but may not correspond to anatomical sex), whereas the gender performance of drag queens typically does not correspond to either gender identity or anatomical sex. The distinction between performance and identity has not been utilized in sociolinguistic research, but it is potentially crucial, for the linguistic manifestations tied to performance are likely to be quite different from those related to personal identity. In identity performance, out-group stereotypes concerning the behavioral patterns of the group associated with the performed identity are likely to be more important than actual behavior or the group's own behavioral norms (Hall 1995). Audience assumptions and expectations may crucially help to coconstruct a performance that successfully conveys a particular identity regardless of the accuracy of the linguistic performance when compared to the behavior of "authentic" holders of the

Indexing Polyphonous Identity

319

identity in question (Preston 1992). Thus the language used in a performed identity is likely to differ from the actual speech of those who categorize themselves as having that identity. In addition to differences in linguistic form, a performed identity and a selfcategorized identity are associated with different social factors. Robert Le Page and Andree Tabouret-Keller (1985) offer four conditions that must be met if one is successfully to match the behavior of groups with which one wishes to identify: (1) identification of the groups; (2) access to the groups; (3) motivation to join them and reinforcement from group members; and (4) ability to modify one's behavior. Although these conditions may be necessary for the creation of identity based on selfcategorization, they may not be required for the creation of a successful performed identity. For the sake of performance, it may be sufficient simply to identify the groups in question and to have the ability to modify one's behavior. One does not necessarily need access to the groups and one certainly does not need motivation to join the groups. In fact, in many cases performance may be used as a means to actually create distance from the group in question (as in the case of blackface). Reinforcement from the group is likely to be absent, and indeed, the performance may cause revulsion of members of the group itself (as with some feminist responses to drag). It is likely that in performing identity, reinforcement from the audience or listener will be more important than the actual behavior of the group being imitated; as with the mocking use of an ethnic dialect, performed identities may actually reflect disdain for the imitated group. Speakers in performance need only adjust their linguistic behavior to the extent necessary to index the identity in question. Such an adjustment may in fact be quite slight, possibly even consisting only of the use of specific lexical items (cf. Preston 1992). Performances by AADQs are often judged (by audiences and other drag queens) on the basis of realness, or the ability to seem to be or "pass" (cf. Bucholtz 1995) as a "real" woman. In order to be "flawless," a drag queen must be "real": Her performance must plausibly lead (usually straight) outsiders to assume that she is anatomically female. Any response (whether reinforcement, rejection, or simply acknowledgment) from actual women is unimportant in the creation of a successful performance. What matters is the response of other gay men and drag queens, who base their judgment not on the actual behavior of women but rather on stereotyped assumptions concerning "feminine" behavior. The performer will also base her performance on stereotypes (and on her assumptions concerning the stereotypes held by the audience). These stereotypes may sometimes reflect the misogynist attitudes and sexist assumptions of the performer. Traditional studies of performance have stressed the performer's responsibility to demonstrate communicative competence before an audience (cf. Bauman 1977; Briggs 1988; Hymes 1981). In drag performances, the performer must be able to produce a "real" feminine speech style or a feminine way of speaking that would sound convincing to someone who did not know that the performer was actually anatomically male. Ironically, the success of drag also depends on making the audience aware that this performance is indeed "false" in some sense (that is, the audience must be reminded that the performer is biologically male). Because a successful drag performance is one in which the audience accepts that the performer could

320

IDENTITY AS IMPROVISATION

pass as a woman, the audience must be occasionally reminded that the performer is indeed performing rather than claiming a female identity. Thus, although glam queens present an external image of exaggerated femininity, they also use language both to create and to undermine this surface image. For example, drag queens frequently use a stereotypically "feminine" speaking style, but a stereotypically "masculine" voice may break through during the performance, creating a polyphonous and often ambiguous performed identity. Within the performances of AADQs in particular, a crucial aspect of communicative competence is the rhetorical device of signifyin(g) (Abrahams 1976; Gates 1988; Mitchell-Kernan 1972; Smitherman 1977). In signifying, the full intended meaning of an utterance does not rest solely on referential meaning. Rather, an utterance is valued because of its ability to index an ambiguous relationship between the signifier and the signified. Thus the signifier does not simply correspond to a particular concept but indexes a rhetorical figure or skill at verbal art. In signifying, a speaker draws attention to language itself, particularly to her or his skill at using language creatively. Specific attention to language (rather than referential content) may be created through a variety of devices, including the creation of polysemy or ambiguity, the creative use of indirection (Morgan 1991), and the contrastive use of a particular style, as in reading dialect (Morgan, chapter 1, this volume). Signifying relies on the listener's ability to connect the content of an utterance to the context in which it occurs and specifically to sort through the possible meanings and implications of an utterance and realize both the proper meaning and the skill of the speaker in creating multiple potential meanings. Successful performances by AADQs typically include cases of signifying. A highly effective instance of signifying is sometimes picked up by other drag queens for use in their own performances. Example (1) has been used fairly widely by various AADQs in Texas: (1)

Drag queen: Audience: Drag queen: Audience: Drag queen: Audience: Drag queen:

Everybody say "Hey!" Hey! Everybody say "Ho!" Ho! Everybody say "Hey! Ho!" Hey! Ho! Hey! How y'all doin?

This example draws on the form of a call-response routine, a rhetorical trope sometimes associated with African American sermons and often used in drag performances. The example relies on the polysemy of the word ho as both an "empty" word frequently used in call-response routines by drag queens and as an equivalent of whore. After leading the audience into the chant and getting them to yell "Hey! Ho," the drag queen reinterprets the word ho, taking the audience's chanting of ho as a vocative. The polysemy is dependent on the connection between the utterance and the context. Performances of AADQs contain numerous examples of signifying, in many of which the polysemy is achieved through the juxtaposition of language styles or social dialects.

Indexing Polyphonous Identity

321

White women's language among AADQs Marjorie Garber (1992) notes that there is a long tradition of simultaneous movement across lines of both gender and race/ethnicity. For AADQs, the move to perform female gender is often accompanied by a simultaneous movement across lines of race and class. Sometimes an AADQ will openly state that she is actually white. For example, The Lady Chablis, a Savannah drag queen made famous in John Berendt's (1994) Midnight in the Garden of Good and Evil, often refers to herself as a "white woman." Berendt describes one of her performances as follows: '"I am not what I may appear to be,' she will say with apparent candor, adding, 'No, child, 1 am a heterosexual white woman. That's right, honey. Do not be fooled by what you see. When you look at me, you are lookin' at the Junior League. You are lookin' at an uptown white woman, and a pregnant uptown white woman at that'" (Berendt 1996:14). As a "pregnant uptown white woman," The Lady Chablis moves from being a gay African American who is biologically male and from a working-class background to being upper-class, white, heterosexual, and female. In her autobiography, The Lady Chablis refers to herself and a close circle of friends as the Savannah League of Uptown White Women (or SLUWW). SLUWW was formed "to honor the belief that all of us [the league members] is entitled to spend our days sitting up under hairdryers, going to lunch, and riding around town shopping—all at somebody else's expense" (1996:173; original emphasis). She defines an "uptown white woman" as "the persona of a classy, extravagant, and glamorous woman—big car, big rings, etc.," adding parenthetically, "(This term can be used for all women regardless of color)" (1996:175; original emphasis). The term white woman refers primarily to a class rather than an ethnic distinction and also collapses the categories of '"real" women' and 'drag queens'. Thus each of us has the potential to become an "uptown white woman," no matter what our sexual, racial, ethnic, or gender identity may be. Instead of suggesting a category based on sex or race, white woman indexes a prevailing ideology of gender, class, sexuality, and ethnicity that enforces a particular view of what constitutes "femininity" in U.S. culture. The combination of particular identity stances (white, rich, female, and heterosexual) works to produce a cultural conception of what constitutes the feminine ideal. This ideal femininity is often associated with the idea of being a "lady." As Esther Newton (1979:127) notes, "Most female impersonators aspire to act like 'ladies,' and to call a woman a 'lady' is to confer the highest honor." The "white-woman" style of speech as used by AADQs represents a stereotype of the speech of middle-class white women, of how to talk "like a lady." This stereotype is closely tied to Robin Lakoff s notion of "women's language" (WL), which also depicts a stereotype of white middleclass women's speech, a fact that Lakoff herself recognized (1975:59). Mary Bucholtz and Kira Hall (1995) have noted the pervasiveness of WL as a hegemonic notion of gender-appropriate language. Because it is such a strong symbol of ideal femininity, WL is a powerful tool for performing female identity. For example, Lillian Glass (1992) reports that she used Lakoff s (1975) Language and Woman's Place in speech therapy with a male-to-female transsexual to produce gender-appropriate language use. Similarly, Jennifer Anne Stevens (1990) presents many of the features of WL in her guidebook for male-to-female transgenders. In addition to discussing issues of hormones

322

IDENTITY AS IMPROVISATION

and offering tips on choosing makeup and clothing, Stevens presents details about creating a feminine voice. Many elements of Lakoff s WL are included in the features of feminine speech that Stevens suggests, including tag questions, hedges, the use of "empty" adjectives, the absence of obscenities, and the use of intensive so. Because of the power of WL as a stereotype of how middle-class white women talk (or "should" talk), I will use it as a basis for discussing the "white-woman" style of AADQs' speech. Here my use of the term white-woman style is intended to reflect this stereotyped representation rather than the real behavior of any actual white women. Lakoff summarizes the main characteristics of WL as follows: 1. Women have a large stock of words related to their specific interests, generally relegated to them as "woman's work": magenta . . . dart (in sewing), and so on. 2. "Empty" adjectives like divine, charming, cute. 3. Question intonation where we might expect declaratives: for instance, tag questions ("It's so hot, isn't it?") and rising intonation in statement contexts ("What's your name, dear?" "Mary Smith?"). 4. The use of hedges of various kinds. Women's speech seems in general to contain more instances of "well," "y'know," "kinda," and so forth. 5. Related to this is the intensive use of "so." Again, this is more frequent in women's than men's language. 6. Hypercorrect grammar: Women are not supposed to talk rough. 7. Superpolite forms: Women don't use off-color or indelicate expressions; women are the experts at euphemism. 8. Women don't tell jokes. 9. Women speak in italics [i.e., betray the fear that little attention is being paid to what they say]. (1975:53-56)

Of these nine elements of WL, AADQs utilize only the first six. Several of these, such as the use of precise color terms and "empty" adjectives, overlap with gay male speech. However, AADQs typically distinguish between the two styles. For example, the "empty" adjectives in the gay-male style of speaking are characteristically "gay," such as flawless, fierce, fabulous, and so on. In the "white-woman" style, the empty adjectives are more similar to those discussed by Lakoff. For example, in (2a) an AADQ asked why I was studying "drag language." When told that I was a linguist, she responded with Oh, really, that's cute, where cute seems fairly devoid of meaning. Example (2b), also from a Texas AADQ, provides further instances of intensifiers and "empty" adjectives (really and cute). (Note also the use of intensive so in example (2b).)3 (2a)

A: . . . drag language? What is ... B: He's a linguist . . . linguistic. A: (overlap) My brain is dead. A: Oh really . . . that's cute. (2b) Oh, my, my . . . I lost a ring y'all and I am vixed [= vexed] Really vixed, because . . . I have no idea where it is and I just bought that little ring and it's so cute.

Indexing Polyphonous Identity

323

Example (3) is taken from an interview with RuPaul on "The Arsenio Hall Show." This example demonstrates the use of final high intonation on declarative sentences (Lakoffs second characteristic of WL): (3) L H L H* L You guys, I wish there was a camera so I could remember H* L H all the love you're sending to me L H and the . . L H the love energy from over here. L H* L You're absolute the best.

In these examples, AADQs use careful, "Standard" English phonology. In other words, they use "correct" prescriptive pronunciations as opposed to phonological features stereotypically associated with AAVE. This "white-woman" style is the most common speaking style among AADQs, and the ability to use this style is considered vital to the success of AADQs' performances. The use of this style also distinguishes AADQs from other African American gay men. Thus it functions both to index stereotypes of white femininity and to construct a unique drag-queen identity that appropriates and reworks the symbols of "ideal" femininity. Performing polyphonous identity Although the use of the white-woman style of speaking is closely tied to ideals of expected feminine behavior, AADQs do not use it exclusively. If such speakers actually wanted to be white, one would expect them to use white women's speech in an attempt to gain the social standing afforded to white women. Frequently, however, they use the "white-woman" style as a type of dialect opposition (Morgan, chapter 1, this volume) in which this style is contrasted with other styles of speaking, primarily AAVE, to highlight social difference. In fact, the use of white women's speech among AADQs is itself a type of signifying. It indexes not only the social status or identity of white women but also the ability of a particular AADQ to use the "white-woman" style effectively. Most of the remaining examples are cases of polysemy created through dialect opposition, reflecting the ambiguity of signifying. These examples demonstrate that although the "white-woman" style is a vital characteristic of AADQs' identity, its use does not imply an underlying desire to be white. Rather, the white-woman style is one of numerous stylistic voices related to drag-queen identity and is used to create specific personas and changing identities throughout the course of a performance. Other stylistic choices, such as AAVE or gay male speech, are used to "interrupt" the white-woman style, to point out that it reflects a performed identity that may not correspond to the assumed biographical identity of the performer.

324

IDENTITY AS IMPROVISATION

As noted earlier, AADQs do not adopt the last three characteristics of WL (avoiding off-color expressions, not telling jokes, and speaking in italics). Although all of the features of WL are related to "acting like a lady," these three are perhaps the most important keys to "ladylike" behavior. Lakoff notes that they may indicate that women realize "that they are not being listened to" (1975:56). One major difference between the "ladylike" behavior represented by WL and the behavior of AADQs is that "ladies" do not make themselves the center of attention, whereas drag queens often do little else. AADQs sometimes flaunt the fact that they do not meet the standard of proper middle-class women's behavior by using obscenities strategically. In example (4), a drag queen points out that she is not supposed to use words like/ac& and shit, accentuating the fact that she deviates from the prescribed linguistic behavior of middle-class white women: (4)

Are you ready to see some muscles? [audience yells]. . . . Some dick? Excuse me I'm not supposed to say that. . . words like that in the microphone . . . Like shit, fuck, and all that, you know? I am a Christian woman. I go to church. I'm always on my knees.

The statement I'm always on my knees is an instance of signifying in that it conveys double meaning. In the context of the utterance, it suggests that the speaker prays all the time. Because it is spoken by a drag queen in a gay bar, however, it also insinuates that she frequently performs oral sex on other men. The failure to have an ideal "ladylike" way of speaking (the use of obscenities) is paralleled in the failure to have appropriate "ladylike" sexual behavior. Here, the white-woman style co-occurs with obscenities that suggest the "falseness" of the performed white-woman identity. By creating two contrasting voices within a single discourse, the performer plays off of the disjuncture between performed ("female") and biographical ("male") identity. In example (5), a Texas AADQ moves from speaking fairly "Standard" English in a high-pitched voice to using an exaggerated low-pitched voice to utter the phrase Hey what's up, home boy to an African American audience member. This monologue occurred in a gay bar with a predominantly white clientele. The switch serves to reaffirm the fact that the AADQ is African American and biologically male while simultaneously creating a sense of solidarity with the audience member to whom it is addressed. (Note: a butt-fucking tea is anything that is exceptionally good.) (5)

Please welcome to the stage, our next dancer. He is a butt-fucking tea, honey He is hot. Masculine, muscled, and ready to put it to ya, baby. Anybody in here (.) hot (.) as (.) fish (.) grease? That's pretty hot, idn't it? (Switch to low pitch) Hey what's up, home boy? (Switches back) I'm sorry that fucking Creole always come around when I don't need it.

Indexing Polyphonous Identity

325

The speaker apologizes with that fucking Creole always come around when I don't need it, but the word creole is pronounced with a vocalized /!/, and the verb come is spoken without the "Standard" English /+s/ inflection. Thus, in apologizing for her use of AAVE (or "creole"), she continues to include features characteristic of AAVE in her speech (just as the apology for using an obscenity in example (4) involved the continued use of obscenities). This helps shape the statement as a form of signifying by implying that what is spoken does not really convey the full meaning of the utterance. The speaker's continued use of AAVE suggests that she has no intention of actually switching totally into "Standard" English (or of totally giving in to the performed white-woman identity symbolized by that variety of English). Unlike the previous examples, example (6) is not typical of AADQs' performances. I include it here because it deals with a complex set of issues revolving around white stereotypes of African Americans. In this example, performed in an African American gay bar, an AADQ uses the "white-woman" style in acting out an attack on a rich white woman by an African American man. Acting out the rape of any woman is a misogynistic act; yet although this misogyny should not be excused, it is important to note that the main impetus for this piece of data is anger concerning the myth of the African American rapist. As Angela Davis has pointed out (1983), fraudulent charges of rape have historically been used as excuses for the murder (by lynching) of African American men. Because it is based on the racist stereotype of African Americans as having voracious sexual appetites, the myth of the African American rapist operates under the false assumption that rape is a primarily sexual act (and not primarily an act of violence). It assumes that all African American men are desirous of white women and are willing to commit acts of violence in order to feed this desire. The fact that this assumption has no basis is especially heightened in the context of African American gay men, who may not be desirous of any women. Nevertheless, the patrons of the bar must continuously deal with the ramifications of the myth of the Black rapist, including unfounded white fears of violence. Lines 1 through 21 present the attack on the white woman, in which the AADQ, in interaction with a male audience member who assists in the scene, uses the "white-woman" style alternating with AAVE as she moves in and out of the persona of a white woman: (6)

1 2 3 4 5 6 7 8 9 10

11

I'm a rich white woman in {name of wealthy white neighborhood} and you're going to try to come after me, OK? And I want you to j u s t . . . I'm going to be running, OK? And I'm gonna fall down, OK? OK? And I'm just gonna . . look at you . . . and you don't do anything. You hold the gun . . . Goddamn- he got practice, [audience laughter] I can tell you're experienced. [The audience member holds the gun, but so that it faces down, not as if he were aiming it] OK hold it.

326

IDENTITY AS IMPROVISATION

12 13 14 15 16 17 18 19 20

21

You know you know how to hold it, don't play it off . . . Hold that gun . . . Shit. . . Goddamn . . [Female audience member]: Hold that gun! That's right fish! Hold that gun! Shit! OK now, y'all, I'm fish, y'all, white fish witch! And I'm gonna be running cause three Black men with big dicks chasing me! [Points to audience member] He's the leader, OK? Now you know I gotta fall, I want y'all to say, "Fall bitch!" [Audience]: Fall bitch! [The AADQ falls, then rises, makes gasping sounds, alternating with "bumbiddy-butn" imitations of the type of music used in suspense scenes in movies and TV shows] Now show me the gun! [The audience member holds up the gun and the AADQ performs an exaggerated faint]

It is interesting to note that the man holding the gun does not "do anything" (lines 78). Despite the AADQ's insinuation that he is "experienced" (line 10), the audience member fails to hold the gun correctly until a woman in the audience yells at him (line 14). The "white woman" pretends that "Black men with big dicks" are chasing her through the park (line 17) and faints on seeing the man with the gun (line 21). Thus, the African American man is basically passive throughout the exchange and the "white woman" reacts primarily based on fear fed by racism. In the remainder of the segment, the corollary to the myth of the African American rapist is presented, the myth of the promiscuity of the African American woman (Davis 1983:182). In lines 22 through 26, the same scene is acted out with an "African American woman" (speaking primarily in a tough, streetwise "bangee girl" style of AAVE) rather than a "white woman." The "African American woman," on seeing the large feet of the man with the gun (which implies he has a large penis as well), consents to having sex with him, saying that the gun is unnecessary (lines 25—26): 22 23 24 25 26

Now this Black fish Black men's running after her . . I ain't no boy] Fuck y'all! Fuck y'all mother fuckers! [AADQ looks at the gun] You don't have to use that baby, I see them size feet. Come on! Come on!

To focus only on the inescapable misogyny of this example is to miss its political complexity. The performance also touches all aspects of the myth of the African American rapist, the racist assumptions concerning both the "pure and fragile nature" of white women as "standards of morality" and the "bestial nature" of African American women and men. In this highly political performance, the drag queen moves in and out of the personas of narrator, director, and actor in the drama she is creating. She performs a variety of identities indexed by a variety of linguis-

Indexing Polyphonous Identity

327

tic styles to undermine a variety of stereotypes and prejudices that are all too familiar to her audience. Conclusion The examples discussed above suggest that the use of white women's speech by AADQs cannot be interpreted as simply reflecting a desire to be white. The femininity associated with speaking like a "white woman" simultaneously indexes a set of class, gender, and ethnic identities associated with the ideology of what constitutes "ideal" feminine behavior. Although the "white-woman" style is sometimes emblematic of status, it is also used in combination with other stylistic choices to highlight a variety of more critical attitudes toward whiteness. Thus the appropriation of aspects of dominant culture need not necessarily indicate acceptance of its dominating force. Rather, this appropriation can serve as a form of resistance (Butler 1993:137). Indeed, in some cases the appropriation of white women's language does succeed in undermining racist and homophobic assumptions associated with the dominant culture. But arguments concerning the misogyny of drag cannot be brushed aside simply because drag is sometimes subversive. Although the examples in this chapter suggest a form of resistance toward racism and homophobia, they do little to call into question the sexism in American society. The performances of AADQs should not be understood simply as "subversive" or "submissive" with regard to dominant hegemonic culture. The polyphony of stylistic voices and the identities they index serve to convey multiple meanings that may vary across contexts and speakers. A full understanding of a phenomenon such as drag requires that we follow the advice of Claudia Mitchell-Kernan and "attend to all potential meaning-carrying symbolic systems in speech events—the total universe of discourse" (1972:166). NOTES

This chapter is dedicated to the memory of Grainger Sanders (1954-1994). Grainger inspired this research and assisted both in collecting data and in shaping my understanding of AADQs. I could never have thanked him enough. Additional thanks to Gregory Clay, Kathryn Semolic, and Keith Walters. An earlier version of this chapter (Barrett 1994) was presented at the Third Berkeley Women and Language Conference. For different but complementary analyses of some of these data, see also Barrett (1995, 1998). 1. The terms transgenderist and transgender(ed) person are often used as umbrella terms for members of these different groups. They are sometimes seen as alternatives to terms with medical connotations, such as transvestite and transsexual. 2. Following community norms for polite reference, I use she to refer to drag queens when they are in drag. 3. Transcription conventions are as follows:

obscured material

[]

text-external information

328

IDENTITY AS IMPROVISATION

{ }

segment removed from data to ensure anonymity

italics

emphasis

H

high intonation (see McLemore 1991)

L

low intonation

H*/L*

pitch accent

(.)

short pause forming separation between words longer pause (more periods indicate greater length)

underlining

material under discussion

REFERENCES

Abrahams, Roger D. (1976). Talking black. Rowley, MA: Newbury House. Ackroyd, Peter (1979). Dressing up—Transvestism and drag: The history of an obsession. New York: Simon & Schuster. Bakhtin, Mikhail M. (1981). The dialogic imagination. Ed. Michael Holquist. Trans. Caryl Emerson & Michael Holquist. Austin: University of Texas Press. (1984). Problems of Dostoevsky's poetics. Ed. & trans. Caryl Emerson. Minneapolis: University of Minneapolis Press. Barrett, Rusty (1994). "She is NOT white woman!": The appropriation of white women's language by African American drag queens. In Mary Bucholtz, A. C. Liang, Laurel Sutton, & Caitlin Hines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 1-14. (1995). Supermodels of the world, unite!: Political economy and the language of performance among African American drag queens. In William L. Leap (ed.), Beyond the lavender lexicon: Authenticity, imagination and appropriation in lesbian and gay languages. Newark: Gordon & Breach, 207-226. (1998). Markedness and style switching in performances by African American drag queens. In Carol Myers-Scotton (ed.), Linguistic choices as social messages. New York: Oxford University Press, 139-161. Bauman, Richard (1977). Verbal art as performance. Prospect Heights, IL: Waveland Press. Beame, Thorn (1983). Racism from a black perspective. In Michael J. Smith (ed.), Black men, white men. San Francisco: Gay Sunshine Press, 57-62. Berendt, John (1994). Midnight in the garden of good and evil. New York: Random House. (1996). Introduction: Chablis and me. In The Lady Chablis with Theodore Bouloukos, Hiding my candy: The autobiography of the grand empress of Savannah. New York: Pocket Books, 12-18. Boykin, Keith (1996). One more river to cross: Black and gay in America. New York: Anchor Books. Briggs, Charles L. (1988). Competence in performance: The creativity of tradition in Mexicano verbal art. Philadelphia: University of Pennsylvania Press. Bucholtz, Mary (1995). From mulatta to mestiza: Passing and the linguistic reshaping of ethnic identity. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 351-373. Bucholtz, Mary, & Kira Hall (1995). Introduction: Twenty years after Language and woman's place. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 1-22. Bullough, Vern L., & Bonnie Bullough (1993). Cross dressing, sex, and gender. Philadelphia: University of Pennsylvania Press.

Indexing Polyphonous Identity

329

Butler, Judith (1990). Gender trouble: Feminism and the subversion of identity. New York: Routledge. (1993). Bodies that matter: On the discursive limits of "sex. " New York: Routledge. Davies, Bronwyn, & Rom Havre (1990). Positioning: The discursive production of selves. Journal/or the Theory of Social Behaviour 20(l):43-63. Davis, Angela (1983). Women, race and class. New York: Random House. DeMarco, Joe (1983). Gay racism. In Michael J. Smith (ed.), Black men, white men. San Francisco: Gay Sunshine Press, 109-118. Duberman, Martin (1993). Stonewall. New York: Dutton. Eckert, Penelope, & Sally McConnell-Ginet (1992). Think practically and look locally: Language and gender as community-based practice. Annual Review of Anthropology 21:461-490. Feinberg, Leslie (1996). Transgender warriors: Making history from Joan of Arc to RuPaul. Boston: Beacon Press. Fleisher, Julian (1996). The drag queens of New York: An illustrated field guide. New York: Riverhead Books. Frye, Marilyn (1983). The politics of reality: Essays in feminist theory. Trumansburg, NY: Crossing Press. Garber, Marjorie (1992). Vested interests: Cross-dressing and cultural anxiety. New York: Routledge. Gates, Henry Louis, Jr. (1988). The signifying monkey: A theory of African-American literary criticism. Oxford: Oxford University Press. Gilbert, Sandra M. (1982). Costumes of the mind: Transvestism as metaphor in modern literature. Tn Elizabeth Abel (ed.), Writing and sexual difference. Chicago: University of Chicago Press, 193-220. Glass, Lillian (1992). He says, she says: Closing the communication gap between the sexes. New York: Putnam. Hall, Kira (1995). Lip service on the fantasy lines. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 183-216. Harper, Phillip Brian (1993). Eloquence and epitaph: Black nationalism and the homophobic impulse in responses to the death of Max Robinson. In Michael Warner (ed.), Fear of a queer planet: Queer politics and social theory. Minneapolis: University of Minnesota Press, 239263. Hilbert, Jeffrey (1995). The politics of drag. In Corey K. Creekmur & Alexander Doty (eds.), Out in culture: Gay, lesbian and queer essays on popular culture. Durham, NC: Duke University Press, 463-469. hooks, bell (1989). Talking back: Thinking feminist, thinking Black. Boston: South End Press. Hymes, Dell (1981). "In vain I tried to tell you": Essays in Native American ethnopoetics. Philadelphia: University of Pennsylvania Press. Kroskrity, Paul V. (1993). Language, history, and identity: Ethnolinguistic studies of the Arizona Tewa. Tucson: University of Arizona Press. Labov, William (1972). Language in the inner city: Studies in the Black English Vernacular. Philadelphia: University of Pennsylvania Press. The Lady Chablis, with Theodore Bouloukos (1996). Hiding my candy: The autobiography of the grand empress of Savannah. New York: Pocket Books. Lakoff, Robin (1975). Language and woman's place. New York: Harper & Row. Le Page, R. B., & Andree Tabouret-Keller (1985). Acts of identity: Creole-based approaches to language and ethnicity. Cambridge: Cambridge University Press. Lurie, Allison (1981). The language of clothes. New York: Random House. MacKenzie, Gordene Olga (1994). Transgender nation. Bowling Green, OH: Bowling Green State University Popular Press.

330

IDENTITY AS IMPROVISATION

Marcus, Eric (1992). Making history: The struggle for gay and lesbian equal rights 1945-1990: An oral history. New York: Harper & Row. McElhinny, Bonnie S. (1995). Challenging hegemonic masculinities: Female and male police officers handling domestic violence. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 217—243. McLemore, Cynthia Ann (1991). The pragmatic interpretation of English intonation: Sorority speech. Ph.D. diss., University of Texas, Austin. Mitchell-Kernan, Claudia (1972). Signifying and marking: Two Afro-American speech acts. In John I. Gumper/. & Dell Hymes (eds.), Directions in sociolinguistics. New York: Holt, Rinehart & Winston, 161-179. Monteiro, Kenneth P., & Vincent Fuqua (1994). African American gay youth: One form of manhood. High School Journal 77(l-2):20-36. Morgan, Marcyliena (1991). Indirectness and interpretation in African American women's discourse. Pragmatics l(4):421-435. Newton, Esther (1979). Mother camp: Female impersonation in America, Chicago: University of Chicago Press. Ochs, Elinor (1992). Indexing gender. In Alessandro Duranti & Charles Goodwin (eds.), Rethinking context. Cambridge: Cambridge University Press, 335-358. Peterson, John L. (1992). Black men and their same-sex desires and behaviors. In Gilbert Herdt (ed.), Gay culture in America: Essays from the field. Boston: Beacon Press, 87—106. Preston, Dennis (1992). Talking black and talking white: A study in variety imitation. In Joan H. Hall, Nick Doane, & Dick Ringler (eds.), Old English and new: Studies in language and linguistics in honor of Frederic G. Cassidy. New York: Garland, 327-354. Raymond, Janice (1994). The transsexual empire: The making of the. she-male. New York: Teachers College Press. (Original work published 1979) (1996). The politics of transgenderism. In Richard Ekins & Dave King (eds.), Blending genders: Social aspects of cross-dressing and sex-changing. New York: Routledge, 215223. RuPaul (1995). Lettin it all hang out: An autobiography. New York: Hyperion. Showalter, Elaine (1983). Critical cross-dressing: Male feminists and the woman of the year. Raritan 3(2): 130-149. Simmons, Ron (1991). Tongues untied: An interview with Marlon Riggs. In Essex Hemphill (ed.), Brother to brother: New writings by Black gay men. Boston: Alyson, 189-199. Smith, Max C. (1986). By the year 2000. In Joseph Beam (ed.), In the life: A Black gay anthology. Boston: Alyson, 224-229. Smitherman, Geneva (1977). Talkin and testifyin: The language of black America. Boston: Houghton Mifflin. Stevens, Jennifer Anne (1990). From masculine to feminine and all points in between: A practical guide for transvestites, cross-dressers, transgenderists, transsexuals, and others who choose to develop a more feminine image . . . and for the curious and concerned. Cambridge, MA: Different Path Press. Tinney, James S. (1986). Why a gay Black church? In Joseph Beam (ed.), In the life: A Black gay anthology. Boston: Alyson, 70-86. Trudgill, Peter (1983). On dialect: Social and geographical perspectives. New York: New York University Press. Walters, Keith (1996). Contesting representations of African American language. In Risako Ide, Rebecca Parker, & Yukako Sunaoshi (eds.), SALSA III: Proceedings of the Third Annual Symposium about Language and Society—Austin (Texas Linguistics Forum 36). Austin: University of Texas, Department of Linguistics, 137-151.

Indexing Polyphonous Identity

331

White, Edmund (1980). The political vocabulary of homosexuality. In Leonard Michaels & Christopher Ricks (eds.), The state of the language. Berkeley: University of California Press, 235246. Williamson, Judith (1986). Consuming passions: The dynamics of popular culture. London: Marion Boyars. Woolard, Kathryn (1995). Gendered peer groups and the bilingual repertoire in Catalonia. In Pamela Silberman & Jonathan Loftin (eds.), SALSA II: Proceedings of the Second Annual Symposium about Language and Society—Austin (Texas Linguistics Forum 35). Austin: University of Texas, Department of Linguistics, 200-220.

17

ANNA LIVIA

"She Sired Six Children" Feminist Experiments with Linguistic Gender I utterly refuse to mangle English by inventing a pronoun for 'he/she'. "He" is the generic pronoun, damn it. —Ursula Le Guin, "Is Gender Necessary?" (1979) I dislike the so-called generic pronouns "he/him/his" which exclude women from discourse. . . . "They/them/their" should be restored . . . and let the pedants and pundits squeak and gibber in the streets. —Ursula Le Guin, "Is Gender Necessary? Redux" (1989)

recurring theme for feminist novelists of the 1970s was the question of pronouns. Feminist theorist Elaine Morgan notes that if you use the pronoun he in a book about the history of man, "before you are halfway through the first chapter a mental image of this evolving creature begins to form in your mind. It will be a male image and he will be the hero of the story" (quoted in Spender 1980:151). Morgan comments that the use of he is "a simple matter of linguistic convenience" which, given the context, is clearly intended ironically, for the use of generic he is neither simple nor convenient. She goes on to remark that she longs to find a book that begins, "When the first ancestor of the human race descended from the trees, she had not yet developed the mighty brain that was to distinguish her so sharply from other species." If the feminine pronoun in Morgan's imaginary textbook produces a sense of shock, this reaction indicates that the reader has, despite the lack of overt markers of masculinity, nevertheless imagined a man. Thus the use of the generic he is no simple matter. Nor is it clear that linguistic convenience would cause one to choose the pronoun he to anaphorize man. This choice is more a matter of gender concord, since singular they is the pronoun of preference in informal speech (where convenience might be said to outweigh elegance) to anaphorize indefinite antecedents: Everyone loves their mother. Most theorists have rejected the substitution of the female pronoun for the male, believing that this would merely turn the problem around without solving it.1 In some cases the epicene (common-gender) but impossibly formal-sounding one is used in preference to generic he or "singular they" the latter still considered 332

"She Sired Six Children "

333

ungrammatical by purists and contemporary linguistic watchdogs.2 Writing in French, Monique Wittig makes the epicene on the hero/ine of L'Opoponax (1964), her first novel, where, in keeping with standard spoken French, the pronoun takes the place of all three persons, both singular and plural, at different times, depending on context. Although this strategy works admirably in French, in English one (on's nearest equivalent) sounds stuffy and scholastic. Writing of on and one, Wittig claims, "There is in French, as there is in English, a munificent pronoun . . . not marked by gender, a pronoun that you are taught in school to systematically avoid" (1986:68). She complains that the English translator Helen Weaver had absorbed the lesson on avoiding one so well that she "managed never to use it" in the translation of L'Opoponax. Wittig admits that English one "sounds and looks very heavy" but, she claims, "no less so in French." This is, however, simply not the case. French teachers tell their pupils not to use on in writing because it sounds colloquial, uneducated, marked for the spoken language. In contrast, in English one is often encouraged in formal writing as having a more impersonal, scholarly, and therefore more authoritative voice. With regard to register, on and one are not equivalents but opposites. Elsewhere, Wittig (1992) employs one as an indefinite pronoun, sometimes inclusive of, at other times in contrast to, the subject pronoun /: (1) (2) (3) (4)

The only thing to do is to stand on one's own feet as an escapee . . . One must accept that my point of view may appear crude . . . First one must step out of the tracks . . . One might have to do without. . .

Although the use of indefinite one in examples (1), (3), and (4) sounds merely oldfashioned and rather aristocratic (to my British ears), the use of indefinite one and the deictic my to indicate the same referent in example (2) is incongruous because instead of forming an integral part of that body of persons encapsulated in one, the first person has been explicitly excluded. In French, on the other hand, the same sentence (On doit accepter que man point de vue paraisse grassier) is acceptable. French on resists the grammatical imperative to provide gender information about the referent, but English one, though equally epicene, is not nearly so versatile, being restricted both by register (formal) and reference (inclusive of speaker).3 Writing of the century-long search for an epicene third-person pronoun, of which he gives more than eighty examples from more than 200 sources between 1850 and 1985, Dennis Baron (1986) dismisses the whole endeavor as "The Word that Failed," the heading for his tenth chapter. Indeed, he states explicitly, "the creation of a common gender pronoun to replace the generic masculine 'he' . . . stands out as the (linguistic reform) most often attempted and the one that has most often failed" (1986:190). This condemnation of failure is interesting. If success is to be measured only by the entry of one of these pronouns into everyday language, then the attempt has indeed failed, although Baron's own impressive list of contenders is testimony to the depth and longevity of concern about the issue. Yet singular they has shown a dogged resistance to the attempts of conservative grammarians to eradicate it, whereas the various invented pronouns such as ne/nis/ner, ho/hom/hos, and shis/'shiny'shims/ shimself exist only in the articles exhorting their use.

334

IDENTITY AS IMPROVISATION

Baron is concerned not with literary uses of epicene pronouns but with their uptake in the language at large. This is, of course, a perfectly reasonable distinction. The carefully selected, gracefully formulated phrases of fiction bear little resemblance to spontaneous speech with its vivid lexicon and distinct syntactic structure. One cannot make valid claims about the spoken language by considering the literary texts produced by its speakers. Nevertheless, the role of literature in introducing, promoting, and popularizing specific words and phrases cannot be dismissed. Where would we be without the malapropism, so called after Mrs. Malaprop of The Rivals, whose every second word was mal a propos (not to the point)? Would the phrase Big Brother is watching you ever have seemed sinister without George Orwell's 19841 As Laurel Sutton makes plain in her contribution to this volume (chapter 8), the planned language of writers can provide useful insights into language ideology (see also Inoue 1994). Let us turn to a consideration of the work of feminist novelists who took up the challenge of the generic masculine, creating ingenious responses to the masculine prerogative in their own novels. In The Kin of'Ata Are Waiting for You (1971), Dorothy Bryant proposes kin, unmarked for either gender or number; in The Cook and the Carpenter (1973), June Arnold introduces na, nan, and naself, in Woman on the Edge of Time (1976), Marge Piercy uses person as subject pronoun and per as object pronoun and possessive. Ursula Le Guin's science-fiction novel The Left Hand of Darkness (1969), which features the generic masculine he for the ambisexual Gethenians, caused such a storm that she was obliged to respond. (Her responses are discussed at length later.) Since three of these novels have been continuously in print for almost 20 years and all four have sold thousands of copies, they can hardly be said to have failed. Each time a reader encounters the neologisms kin, na, or per, she or he is obliged to grapple with the ideological motivation behind these terms: Why have these pronouns been invented? What is wrong with the traditional pronouns they replace? What purpose do the neologisms serve? What effect do they have? Against a 1970s backdrop of identity politics, in which gender was a crucial ingredient, what did it mean to withhold gender information? Insofar as the aim of each author was to raise these questions, the neologistic pronouns work and keep working each time the books are read. Baron uses a quotation from Mary Daly as an epigraph to his chapter on epicene neologisms: "It is a mistake to fixate on the third person singular." In the context of a chapter entitled "The Word that Failed," this quotation might mean that it is a mistake to be obsessed with the generic he, since there are other, more important problems to be solved in the feminist struggle. However, in the original text from which the quotation comes, Daly continues by referring to Monique Wittig's description of how the first-person singular excludes women: "T [Je\ as a generic feminine subject can only enter by force into a language which is foreign to it, for all that is human (masculine) is foreign to it, the human not being feminine grammatically speaking but he \il\ or they [Us]" (Wittig quoted in Daly 1979:18-19). Daly's point, then, is not that it is a mistake to point out the sexism of pronouns but that the thirdperson singular should not be singled out for criticism.4 It is an important point, because all the writers considered in this chapter aim to highlight, each in different ways, the linguistic derogation of women and to redress the balance to some extent by the effect their fiction has on the reader. None "fixate on the third person singular." As Le Guin recalls in the introduction to the 1976 edition of The Left Hand of Darkness, in the mid-1960s she was caught up in a "groundswell of feminist activ-

"She Sired Six Children "

335

ity" and felt a need to define for herself the meaning of gender and sexuality. This was a period in which women were questioning all their previous assumptions about the relationship between the sexes and the role of women. Consciousness-raising groups were mushrooming, and new solutions to old problems such as childbearing, child rearing, and family life were discussed with extraordinary vigor and energy. It was in this political climate that Le Guin wrote The Left Hand of Darkness (1969), a novel set among the androgynous people of Gethen whose bodies, for two-thirds of each month, are ungendered. During the last third of the month they enter "kemmer," the human equivalent of being in heat. Female or male genitalia develop when they come in contact with another Gethenian in kemmer, but they never know in advance which genital formation they will exhibit. A Gethenian who has borne three children may sire two more, for example. As the quotations that head this chapter show, Gethen brought with it the thorny problem of which pronouns to use to refer to such sexless/ duosexual beings. Le Guin's initial solution was to use the masculine generic. In the novel an Investigator from the Ekumen (an observer with both a political and an anthropological function) voices the author's views on pronouns. As in any work of fiction, author and narrator are separate entities, often with distinct or opposing worldviews and their own ethos. In the following passage, however, Le Guin's 1976 view of the role of the generic masculine (seen in Le Guin 1979 as well) may be heard in the voice of the narrator/Investigator: "You cannot think of a Gethenian as 'it'. They are not neuters. They are potentials or integrals." Instead, the Investigator asserts, "I must use 'he', for the same reasons as we used the masculine pronoun in referring to a transcendent god: it is less defined, less specific, than the neuter or the feminine" (1976:70). The Investigator remarks further that the use of the masculine generic pronoun leads her to think of Gethenians as men rather than as menwomen. However, her presumption of masculinity goes beyond the ambisexual Gethenians, for when she speaks of the as-yet-unappointed envoy from the Ekumen, she again uses the masculine pronoun. She assumes the envoy will be male, predisposing readers of her dispatch to appoint a man. This problem apparently does not exist in the language of the Gethenians. We are told that they rejoice in a "human pronoun" that encodes information regarding number and animacy but not gender. In another telling passage, Le Guin shows that she is at some level aware that masculine pronouns prompt a masculine reading. In the opening scene, the envoy from the Ekumen reports on a parade and introduces an unknown member of the crowd as "the person on my left." The character is subsequently referred to as this person, the demonstrative this providing a referential link to the first mention. In the next sentence the anaphoric his is used to refer to the person's forehead: "Wiping sweat from his dark forehead the man—man I must say, having said he and his . . ." (1976:11). If the pronoun he had truly been generic, there would have been a choice of feminine or masculine designation, but the envoy insists that he has no choice but to call the person a man; only men may be referred to as he. Le Guin was roundly criticized by feminist readers and reviewers in the late 1960s and the 1970s for her use of the generic masculine. In response to this criticism she published "Is Gender Necessary?" (1979), the article from which the quotation that opens this chapter comes. She argues that while the generic masculine is problematic, it is less restricted than either the feminine she or the neutral it and far preferable to the invention of clumsy neologisms. The story, however, did not end with

336

IDENTITY AS IMPROVISATION

this statement, and neither did feminist criticism. The groundswell of feminist activism launched itself on the language. By the mid-1970s, the French Mouvement de Liberation des Femmes was grappling with the importance of semiotics and other signifying systems to the construction of the concept 'woman'. Their American cousins, having taken upon themselves the task of reassessing the world, the history of consciousness, and the individual psyche from the new feminist perspective of a woman-centered ontology, followed in the footsteps of the French and turned to an analysis of ideology and its linguistic base. In reaction, Le Guin, being part of this new period of activism, produced a piece of informed self-criticism in an essay written in 1989 (from which comes the second quotation that heads this chapter). This essay would be interesting for its structure alone—a sous-rature approach which preserves the original 1979 essay but adds a commentary bracketed in italics. This procedure allows the reader to see the development in Le Guin's thinking, a reminder that one had to go there to get here. The 1989 voice argues with its earlier self. The earlier voice states that pronouns were really not the issue and that had she made the female side of the Gethenians more prominent, these androgynous beings would not have struck readers as purely masculine. The later voice turns this statement around: "If I had realized how the pronoun I used shaped, directed, controlled my own thinking, I might have been cleverer" (1989:15). The 1989 voice has absorbed contemporary radical feminist and social-constructionist reworkings of the Sapir-Whorf hypothesis and applied it to her fiction.5 Le Guin's commentary poses the thorny question: What can a novelist do when she realizes 20 years later that she disagrees with herself, especially when her former position has been quoted "with cries of joy" (1989:7) by adverse critics? She can write a public recantation, as Le Guin did, but it was too late to revise the novel itself. It was definitely too late to revise it. However, Le Guin found an ingenious twopronged literary solution to the problem of having one's past remain alive in one's present. Her first response was to reprint "Winter's King" ([1969] 1975), a story that was set on the planet of Gethen and that had actually predated The Left Hand of Darkness. This time, however, she used feminine pronouns throughout, although she was (in 1975) still adamantly opposed to invented pronouns. To refer to indefinite antecedents, Le Guin retained the use of the generic masculine. The semantic clashes that result from the unvarying use of feminine pronouns in "Winter's King" are highly amusing, especially when juxtaposed with those caused by the use of masculine nouns and pronouns in The Left Hand of Darkness:

The Left Hand of Darkness (1969) My landlady, a voluble man (38) the king was pregnant'' (73) a paranoid pregnant king and an egomaniac regent (82) it certainly was difficult to imagine him as a young mother (85)

"Winter's King" (1975) the young king had her back against a wall (94) the ex-king of Karhide knew herself a barbarian (110) "Prince Emran is well. She is with her attendants" (99) She sired six children (117)

"She Sired Six Children "

337

In the novel men get pregnant; in the short story women sire children. It is hard to believe that these early exercises in gender-bending and blending, a sexual mix and match, provoked only righteous anger when their comic aspect is so apparent. The laughter produced by the clash between the generic femininity or masculinity of these examples and the biological facts and cultural identities usually restricted to the other sex points to the subversiveness of Le Guin's pronoun choice (see also Barrett, chapter 16, this volume). The second solution Le Guin devised was to introduce invented pronouns in the 1985 screenplay of The Left Hand of Darkness, although she remarks that these were modeled on a British dialect and were therefore not entirely her creations (1989). The pronouns in question, a/un/a 's, replacing she/her/hers or he/him/his, were accepted quite happily when read aloud, although some members of the audience to whom she presented this solution commented that the subject pronoun a sounded too like a Southern American /. Even here, however, Le Guin is still uneasy with the use of grammatical neologism. Although these pronouns may be acceptable in a screenplay, a guide for oral performance, they would, she hypothesizes, "drive the reader mad in print" (1989:15). This distaste for written neologisms is interesting in an author famous for her contribution to science fiction, a genre that revels in neologism. Indeed, her 1985 novel Always Coming Home includes a twelve-page glossary of invented vocabulary of the Kesh, an imaginary native Northern Californian culture. Le Guin's 20-year struggle to find appropriate pronominal anaphors for her anatomically unique Gethenians and her changes of heart and ideology attest to more than a commitment to feminist endeavor on the part of the novelist. The difficulty of the task itself demonstrates that, in contrast to the creation of lexical neologisms so frequent in science fiction, messing with a morphosyntactic staple such as the pronominal system brings with it problems that will affect the whole structure of the discourse. Although Le Guin was the earliest of the writers in this chapter to grapple with questions of language and gender, by the mid-1970s several American novelists participating actively in the women's liberation movement were putting feminist ideas to work in their fiction. For example, in the language of Dorothy Bryant's (1971) Kin of Ata, kin is the only pronoun. Although inanimate objects are carefully divided into feminine and masculine, heterosexually paired in all arrangements from building to planting to eating, there are no pronouns for 'she' and 'he'. There are words for the concepts 'woman' and 'man', but these are seldom used, and there is only one pronoun for all human beings, which makes no distinction of gender or number. So far this is not particularly unusual for a pronoun (after all, English you and, as we have seen, French on encode neither number nor gender). Speakers use the Atan pronoun both in the second person and the vocative, however, and in the third person "they referred to one or more people by it" (1971:50). It was used "the way most people use brother" (50), except that brother is specifically masculine and singular. Bryant provides no examples of Atan, explaining that it has no written form. Bryant's description of this pronoun's properties provides an insight into feminist views of the pronominal ideal. In the language of the Kin of Ata, it is possible to distinguish women from men at the lexical level, but the pronominal system distinguishes only between the animate and the inanimate. Not only are gender and number elided, but so are second and third persons (first-person reference is not men-

338

IDENTITY AS IMPROVISATION

tioned). The second person is deictic—that is, the identity of the referent may be derived only from the context of utterance. The third person, in contrast, classifies the referent as a nonparticipant in the discourse (Benveniste 1966a, b). Kin would thus collapse two of the categories that have been perceived as fundamental to the pronominal function. How such a system might work is hard to imagine. Its promotion as an ideal of human communication is, however, instructive. A pronoun that did not change according to person, gender, or number would presuppose an extremely homogeneous community where such distinctions were mere nuances that could easily be discerned from context. The pronominal function itself would be reduced to that of a placeholder for the verb, a sign indicating "verb coming." Aspects of kin are to be seen in the pronominal systems devised by other novelists. Although kin itself is unique to Bryant's work, ideological homogeneity, egalitarianism, and the absence of sex-role segregation were ideals dear to June Arnold and Marge Piercy. Whereas Bryant merely describes the pronoun of Ata, both Arnold and Piercy employ their epicene neologisms in their respective novels, allowing us the opportunity of a more extensive analysis. They do not, however, banish the traditional pronouns but provide a complex interplay between the two systems. The moments when the systems combine, or collide, are of great significance in each book. The epicene pronouns must be considered in terms of the whole pronominal system of each novel because they cause a redrawing of the distinctions made by all other pronominals. Whereas Le Guin opposed the introduction of grammatical neologisms because their unfamiliar forms render them so noticeable, Arnold and Piercy use this foregrounding effect to emphasize their points. Arnold's The Cook and the Carpenter (1973), unlike the novels previously discussed, is set not on a science-fictional world but in a large town in Texas. The choice of setting removes ideas of gender and language from the arena of the fantastic and places them in the dust of everyday life. This move toward realism closes down, to a certain extent, the number of possible meanings of the text. Indeed, parts of Arnold's story are based closely on actual events that took place during the occupation of a disused police building at 330 East Fifth Street in New York City from January 1 to January 13, 1971, by a group of radical feminists. (The location is now a parking lot for police officers of the Ninth Precinct.) Arnold's story begins as a woman comes to warn a group of people who have just moved to town that local residents plan to throw them out. From the opening line, Arnold introduces the pronouns na and nan without any preamble: "'You know Texas. Do you think it's true?' the cook had asked an hour ago. The carpenter's answer was forgotten in nan pursuit of truth." Nan slips in without explanation, leaving readers to make of it what they may. This strategy is intended to naturalize the pronouns, so that the lack of gender distinctions appears as an unproblematized status quo. As we will see further on, however, even Arnold found the neologisms unwieldy and turned to other tactics to avoid gender disclosure. Three-quarters of the way through the novel, the group occupies a school and plans to use its ample rooms to provide health, education, and other services to the local population. Inevitably, the police come and arrest the occupiers. During the scene of the arrests, epicene pronouns prevail. In the following passage, for example, it is impossible to tell by grammatical means the gender of the deputies or Leslie: "A hand

"She Sired Six Children"

339

covered Leslie's nose and mouth, pushing into nan face; one deputy easily dragged na to the car; another followed by the side, whacking Leslie's body wherever nan stick could land" (1973:139). However, after the group has been thrown into holding cells for the night, the traditional third-person pronouns assert themselves and the neologisms are discontinued: "The sergeant pushed Three to the floor and, with his foot on her back, told the policewoman to take off her shoes, pull downier pants" (139). The questions of why the epicene pronouns are used and why traditional pronouns take over are thus planted implicitly in the text. The reader discovers that the group members are in fact all women, since all now have female pronouns assigned to them. It is by the telling phrase with his foot on her back that everything changes. The nongendered pronouns cannot hold out against such an onslaught; the men have won. Gender is indeed central if it means one sex has its foot on the other's back. The epicene pronouns are backed by a battery of epicene noun phrases: a person, for example, is resumed as the speaker and the white-haired one (7-8), the women's liberation movement is referred to as people who share the group's politics (45); men as those others (61). Occasionally this usage sounds a little stilted: "You let that one look, now let me" (35). Instead of that one a pronoun would seem more natural: "You let her/him/na look, now let me." But the traditional pronoun she or he would, of course, reveal the gender of the referent, and only members of the group occupying the school use the epicene pronouns in their own speech; it is a mark of in-group status, which the narrator shares but the townsfolk do not. To use the neologisms na and nan is to place oneself inside the women's group, to declare one's identification with the feminist ideology. Because readers, in order to understand the text, are obliged to make the conceptual link between separatist practice and gender-neutral pronouns, they too are required to identify with the women's group, or at least to read from the women's position. We later discover that the person who says, "You let that one look," insisting with prurient interest on seeing a diagram of female genital organs, is the male deputy sheriff, to whom epicene pronouns are unknown. The connection between the use of epicene pronouns and the possession of a feminist consciousness is underlined by an incident during which the status quo is overturned. It occurs early on in the novel, when the women's group is still flourishing unhindered and the epicene pronouns prevail. One day, while the children are playing, their cries become highly charged and sex-specific. At this point, instead of the epicene term children, they are referred to as girls and boys: "I am a tiger!" a boy cried. "I am a lion!" a second boy said. . . . "I'm an eagle!" a girl's voice shouted. "Goose!" a boy's voice threw back. (55)

This game quickly degenerates into name-calling: "Bitch!" "Bitch!" . . . "Pussy!"

340

IDENTITY AS IMPROVISATION "Pussy!"

"Cock!" The small female voice grew weaker and shriller; the small male voices boomed like those of ordained men. (56)

Even in the women's group, sexual divisions come to the fore when the little boys begin to act like grown men. It is then not possible to refer to them as an undifferentiated group of "children." This scene makes apparent on the linguistic level conflicts that have also begun to emerge on the political level, as seen in the fraught relations between the women's group and the outside world. The women are trying to bring up nonsexist children in a nonsexist environment, of which the nongendered pronouns are an integral part. But the women cannot control all aspects of the children's lives any more than they can control the land they have occupied. The group exists within, and only with the tolerance of, a wider sexist society whose values and hierarchies also exert an influence on the children and their speech patterns. This is evidenced by the ease with which the children divide themselves into a warring group of boys versus girls and their immediate appropriation of the sexed terms pussy and cock. But what is the metamessage of the book? Is Arnold implying that in a group composed only of women, gender is meaningless, or that police aggression forces a gender split? Because the members of the group are called by nontraditional or epicene names like Nicky, Stubby, and Chris, and their roles are not stereotypical for women— in the opening scene, for example, the carpenter is sanding the porch, and later in the book the cook and the carpenter become lovers—some readers may be surprised when they learn that all are in fact women. Are gendered pronouns unnecessary if sex differences are apparent from subtle cultural clues? Different readers will interpret the novel in different ways, although its frame of social realism, its setting in small-town Texas in the mid-1970s, and its feminist slogans do make a radical-feminist reading more compelling than others. The neologisms are so eye-catching and incongruous, given the realism of the rest of the text, that an obvious interpretation would be that they are intended to focus the reader's attention on what is missing, on the pronouns they replace and the reasons for grammaticalized gender distinctions. The author's own stated purpose in using neologisms may be helpful in this regard. Arnold prefaces her novel with the following statement: Since the differences between men and women are so obvious to all, so impossible to confuse whether we are speaking of learned behavior or inherent characteristics . . . the author understands that it is no longer necessary to distinguish between men and women in this novel. I have therefore used one pronoun for both, trusting the reader to know which is which.

This statement would seem to provide the necessary key to the text. If the neologisms work, if readers can tell who is female and who is male without having this information grammatically coded, then one must conclude that the differences between women and men are indeed obvious and that further coding is unnecessary. It", on the other hand, readers are surprised by the return of the gendered pronouns and the sexual identities they reveal, then sexual difference may not be taken for granted.

"She Sired Six Children"

341

Readers who discover only by the use of she and he two-thirds of the way through the novel that the group that occupies the school is composed of women, whereas the people who oppose them are men, demonstrate by this discovery that gender cannot be assumed. From this perspective, the preface is deliberately ironic. Even without the preface, however, this point is made in the novel itself. The epicene speech of the women's group is utopic; it cannot sustain the onslaught of male violence to which it is subjected. The language is defeated at the same moment as its speakers. Although the book itself may end with the defeat of the women's cause, its publication points to a triumph of a different kind. The Cook and the Carpenter was published by one of the earliest feminist presses in the United States, Daughters Inc., a company June Arnold founded in 1972 to publish books by and for women. Arnold has in fact argued for the exclusion of men from the whole printing and publishing process: "we should wear headbands which state: My words will not be sold to 'his master's voice'" (1976:24). While Daughters Inc. flourished, she maintained a separatist preserve. From this perspective, the great pronoun war turns out to be merely a battle fought on many fronts, a strategy among others for creating and maintaining women's autonomy. The epicene pronouns Marge Piercy created in Woman on the Edge of Time (1976) operate rather differently from those of Arnold. The book is an ingenious mixture of science fiction and social realism. Connie, a twentieth-century psychiatric patient, moves back and forth between her own time zone and the twenty-second century. Changes in third-person pronoun use follow this movement: The narrator and characters in the present time use traditional pronouns, whereas characters in the future time use the epicene neologisms person andper. Unlike in Arnold's novel, in which the characters' gender is not linguistically given until late in the narrative, Piercy's futuristic characters are introduced by the narrator with the pronoun appropriate for their gender. The interchange between the time zones in Piercy's novel is much more fluid than that between Arnold's woman-only and mixed communities, and it causes frequent shifts in the pronominal system. During the sequences set in the future society of Matapoisett, the traditional and epicene pronominal systems coexist: Events described by the future-time characters feature the epicene system; those described by the narrator feature the traditional system. When a character from the future time, Erzulia, takes on the personality of another future-time character, Jackrabbit, and dances like him, the narrator recounts: "She danced Jackrabbit. Yes, she became him. . . . Bolivar's head slowly lifted from his chest. Suddenly Erzulia-Jackrabbit danced over and drew him up. . . . Bolivar began to dance with him/her. The music ended. . . . Bolivar jumped back. 'But I felt per!' he cried out" (1976:316; emphasis added). The use of the traditional pronouns shows the scene from the point of view of Connie, the time traveler. Thus the gender shock of she became him is very apparent since the phrase itself defies the system and a new pronoun, her/him, must be invented. Bolivar's use of the epicene per shows up the world of difference between his community and Connie's. The language of the more advanced time forces changes in that of the earlier century. Accompanying the neologistic pronouns are many epicene lexical items that refer to artifacts and concepts specific to the society of the future: mems (family mem-

342

IDENTITY AS IMPROVISATION

bers, although there may be no biological link among them); kidbinder (a person who looks after other people's children); pillow friends (people with whom one has a sexual relationship). This gender-neutral vocabulary causes Connie frequent difficulty since she speaks a language that encodes sex differences. She notices "people who must be women because they carried their babies on their backs" (71), guessing at gender from cultural information regarding her own time's sex roles rather than those of the future society. Because proper names are frequently epicene too, Connie manages as best she can: "The tall intense person was staring at her. Jackrabbit, Luciente had said: therefore male. He . . ." (77). Connie cannot tell Jackrabbit's sex simply from looking at him, hence the neutral person; she remembers that his name is Jackrabbit and is relieved to settle her dilemma with the masculine pronoun that would be appropriate to refer to a male—or "jack"—rabbit. Another possible future is glimpsed by the reader when Connie tries to find her way to Jackrabbit and Luciente's time but gets lost in Gildina's time. Whereas the society of Matapoisett is an egalitarian Utopia, Gildina's life in a futuristic New York is a nightmare dystopia. Luciente explains to Connie that in Matapoisett "we've reformed pronouns" (42), but in Gildina's version of the future people still use the traditional gendered forms, which represent the extreme pole of a sexually segregated and sexist culture. Occasionally the characters in Matapoisett use gendered pronouns: "I have a sweet friend . . . and her tribe is Harlem-Black. But if you go over you won't find everybody black-skinned like her and me" (103).7 There seems to be no textual reason for this; it answers no necessity of the plot. Thus one is inclined to conclude it is simply a mistake. Indeed, Piercy does seem at times unhappy with the use of epicene pronouns and often avoids them, employing other techniques such as repetition of proper names or ellipsis. This is particularly noticeable in Arthur of Ribble' s memorial speech for his dead son, Jackrabbit (309-310). The passage is characterized by a high number of ellipses and proper names and by relatively little use of pronouns, although there are only two animate referents in the discourse (Arthur and Jackrabbit) and only one, Jackrabbit, who might be referred to in the third person. Because the whole episode is a eulogy to the recently dead Jackrabbit, this referent should remain active throughout the discourse, yet his proper name is given seven times, often where there is no intervening matter or where none of the intervening matter is of the type to deactivate the referent (in other words, there are no references to other characters). The continual use of Jackrabbit's name instead of a pronoun is therefore doubly marked. Connie, on the other hand, the foealizer of the story, is often referred to by a subject pronoun even where her name does not appear in the passage. In one notable episode (194—195), Connie is referred to twelve times as she or her, although her proper name is not given at all. Because she is the vehicle of the plot, the narrator may safely assume she remains active in the reader's mind throughout the novel, but Jackrabbit is just as salient in the scene of his memorial service. It seems that, like Le Guin, Piercy finds morphosyntactic neologisms cumbersome to use, even when they are her own invention. Although the strategy of creating epicene pronouns is the same in both Arnold's and Piercy's novels, the intentions of the two authors are quite different. Piercy intends her work to reflect the egalitarian future world of Matapoisett, a world in which sex-

"She Sired Six Children "

343

role differentiation has been reduced to the point where fetuses are brought to term in giant "brooders," where three mothers, of either sex, volunteer to rear the baby, and where men are able to lactate just as women do. Indeed, Piercy affirms, "I use the common gender pronouns to reinforce the egalitarian nature of that society" (personal communication, December 13, 1993). In marked contrast, gender distinctions are not an issue in Arnold's short-lived women's community because, as a separatist environment, it has only one gender present. Although the children of the women's community comprise both girls and boys, they play a minor role in the novel and are distinguished by sex only in the brief episode analyzed above. It is significant that the children's genders are revealed only in a moment of explicit hostility between the girls and the boys. With Suzette Haden Elgin we move from the 1970s to the 1980s, by which time feminist ideas about language were well established and there was already a substantial literature on the subject of language and gender. Whereas June Arnold and Marge Piercy are more interested in tackling grammatical gender and the hierarchy it creates, Elgin is concerned with lexical semantics and the expression of women's perceptions. In her science-fiction trilogy Native Tongue (1984), The Judas Rose (1987a), and Earthsong (1994), Laadan, the language of women's perceptions, plays a crucial role. Elgin (1987) relates her inspiration for the Laadan trilogy to the ideas she got from reviewing Cheris Kramarae's Women and Men Speaking for the linguistics journal Language. She combined the hypothesis that existing human languages are inadequate to express women's perceptions with the French feminist concept of ecriture feminine and Godel's theorem, as interpreted by Douglas Hofstadter (1979). In Native Tongue, Elgin cites two extensions of Godel's theorem into language and culture: "For any language there are perceptions which it cannot express because they would result in its indirect self-destruction" and "For any culture, there are languages which it cannot use because they would result in its indirect self-destruction" (1984:145). Elgin's trilogy presents a society in which women are treated like children and where the specialized class of Linguist women surreptitiously creates its own language in order to cause unbearable strain on the dominant men's language. The Linguist women plan to spread their language to all women on the planet because once women have linguistic Encodings with which to express their miserable situation, the situation itself will inevitably alter. An Encoding, the Laadan Manual explains, is "the making of a name for a chunk of the world that. . . has been around for a long time but has never before impressed anyone as sufficiently important to deserve its own name" (1984:22). Elgin would appear to be taking direct issue with J. L. Austin's cozy assertion, "Our common stock of words embodies all the distinctions men have found worth drawing, and the connections they have found worth making in the lifetime of many generations" (1961:182). Or rather, Elgin's fiction points out that Austin must be using the term men in its masculine, nongeneric sense, for a multitude of distinctions and connections that women might wish to make are missing from the English lexicon. Elgin remarks that the trilogy revolves around the idea that "the Sapir-Whorf hypothesis is true in its weak form, which means that language does become a mechanism for social change" (1987b:178). It is important to note that she does not say that language limits perceptions—that we are unable to think beyond the confines of our native tongue—but that the unique perceptions of women have no place in the

344

IDENTITY AS IMPROVISATION

well-trodden grooves of a patriarchal language. Because the trilogy, like Le Guin's Left Hand of Darkness, is a thought experiment, not scientific research, it might seem pedantic to ask which women's perceptions form the basis for Laadan: old, young, black, white, lesbian, heterosexual. Elgin's is a science-fictional world where, apparently, these differences are not as salient as in our own. The Linguist/nonLinguist distinction does have enormous importance in the novel, and Laadan is, in fact, the language of the Linguist women's perceptions. (The creation of a world in which linguists are economically and politically the most important group provides a certain gratification for those of us who feel our discipline is more often slighted than celebrated.) The majority of the texts discussed here date from the late 1960s and the 1970s, the period when the women's liberation movement was at its height, and the goal of creating equality between women and men seemed attainable, if not overnight, then at least before the century was over. Literary experiments with gender have, in recent years, tended to take a different trajectory. With the rise of the gay liberation movement and the increased visibility of the gay community, sexual orientation has been added to the simple binary of feminine/masculine. The appearance of increasing numbers of transsexuals and their mobilization as a socially influential group have caused that binary to fragment into a multidimensional prism. Instead of creating epicene neologisms to force the reader to think in terms of persons rather than women and men, novelists such as Anne Garreta (1986), writing in French, and Jeanette Winterson (1993), writing in English, have adopted the strategy of removing all grammatical signs of gender in order to present androgynous characters who transcend sexual distinction. A tactic used in autobiographies by transsexual writers (Noel 1994; Stephens 1983) is to alternate between portrayals of themselves in the feminine and in the masculine to demonstrate the fluidity of gender. As gender has come to be seen less as a fact of individual identity and more as a homogenizing force, as a performative in the Austinian sense, the quest for gender-neutral pronouns has given way to a concept of gender as a social attribute separate from the speaker or the referent. *She sired six children is no longer semantically incongruous, and the disapproving asterisk may be removed, for many male-to-female transsexuals have sired children as men whom they now take care of as women. The sexual, racial, and class identities that formed the basis of many of the most important political platforms of the 1970s and 1980s have given way in the more nuanced 1990s to an acceptance of hybridity, fluidity, and performativity. The class of women, an identification so fiercely fought for in the 1970s, has been undermined from within and without. Fragmentation from within began immediately, as the concept 'woman' was exposed as both an ideological construct and a cultural performance by theorists who themselves had been trained in feminist scholarship (see for example Butler 1990 and Fuss 1989). The "woman-identified woman" began to blur ever more rapidly as male-to-female transsexuals declared that they too identified as women and challenged "women-born women" to justify the exclusion of transsexuals from all-female events (as the annual confrontations at Michigan Womyn's Music Festival bear witness; see off our backs and Transsexuals News Telegraph's letters columns for July, August, and September 1993, 1994, and 1995). The novels

"She Sired Six Children "

345

discussed in this chapter could not be written with a 1990s understanding of the limitations and contradictions of a politics based on identity. The utility of an examination of these earlier feminist texts and their experimentation with pronominal gender lies in the constant slippage that is discovered even within the relatively rigid 1970s concept of identity. By eliminating gendered pronouns, these novels force the reader back on assumptions and presuppositions that may not be borne out by the text. In this way, they expose the fraught and transitory links between the pronoun and the referent and all the cultural baggage of she and he. NOTES

I am grateful to Ursula Le Guin, Marge Piercy, and Dorothy Bryant for providing me with information, references, and opinions about their work and to Roberta Arnold for providing the same assistance with her mother's work. They are, of course, in no way responsible for the use I have made of the material they gave me. This article is an excerpt from the sixth chapter of my dissertation (Livia 1995). 1. For an overview of these discussions see, for example, Betty Lou Dubois and Isabel Crouch (1987), Fatemeh Khosroshahi (1989), and Michael Newman (1992). See also the preface to Deborah Cameron (1985:vii) for the opposite view: "Most sex-indefinite and generic referents in this book will be she and her. If there are any men reading who feel uneasy about being excluded, or not addressed, they may care to consider that women get this feeling within minutes of opening the vast majority of books, and to reflect on the effect it has." It is interesting to note that whereas the early editions (1945-1968) of Dr. Benjamin Spock's Child and Baby Care systematically referred to the baby as he, in response to feminist criticism, pronominal reference since 1976 has alternated between the feminine and the masculine. The baby is no longer a generic boy; instead, each situation presents a specific case: "Every time you pick your baby up ... every time you change him, bathe him, feed him . . . he's getting a feeling that he belongs to you" (1946:3; emphasis added), versus "Every time you pick your baby up—let's assume it's a girl— . . . every time you change her, bathe her, feed her . . . she's getting a feeling that she belongs to you" (1976:2; emphasis added). 2. Dennis Baron (1986:194) reports that singular they had begun to be attacked by purists as early as the eighteenth century. He quotes the example of grammarian Lindley Murray, who insisted on correcting "Can anyone, on their entrance into the world, be fully secure that they shall not be deceived?" to "Can anyone, on his entrance into the world, be fully secure that he shall not be deceived?" Baron also cites the nineteenth-century Quaker grammarian Goold Brown, who declared the sentence "No person should be censured for being careful of their reputation" incorrect because "the pronoun their is of the plural number, and does not correctly represent its antecedent noun person, which is of the third person, singular, masculine." Brown gives no reason for interpreting the noun person as masculine, except that he wants to justify the use of he to anaphorize it. 3. The systematic use of on as the principal pronominal vehicle of L'Opoponax creates other tensions at the discourse level, involving such textual features as focalization, point of view, and narrative voice, as I argue elsewhere (Livia 1995, 1998). For a discussion of other French approaches to the problem of linguistic gender (for example, in the work of Anne Garreta), see Livia (1994, 1995). 4. In languages in which gender is not coded morphosyntactically, such as Finnish, Hungarian, or American Sign Language, the question of creating gender-neutral pronouns

346

IDENTITY AS IMPROVISATION

does not of course arise. This does not mean that these languages do not distinguish between the genders nor that the derogation of women does not occur at some other level, whether lexical, semantic, or pragmatic. 5. For feminist renderings of the Sapir-Whorf hypothesis see for example Dale Spender (1980) and Julia Penelope (1990). Sec also Caitlin Hines (chapter 7, this volume). 6. This quotation is so remarkable that it has made it into Bartlett's Book of Quotations. 1. Matapoisett might be described as a racial Utopia, for the people who inhabit it are from many different racial and ethnic backgrounds, including Native American, Latino, Asian, and African, and there are no signs of interethnic or interracial hostility. However, a more fitting image than a Utopia would be that of a melting pot, for the races are so blended together that little is left of their culture of origin. REFERENCES

Arnold, June (1973). The cook and the carpenter. Plainfield,VT: Daughters Inc. (1976). Feminist presses and feminist politics. Quest: A Feminist Quarterly 3(l):18-26. Austin, J. L. (1961). A plea for excuses. In Austin, Philosophical papers. Oxford: Oxford University Press, 175-204. Baron, Dennis (1986). Grammar and gender. New Haven, CT: Yale University Press. Benveniste, Emile (1966a). La nature des pronoms. In Benveniste, Problemes de linguistique generate. Vol. 1. Paris: GalHmard, 251—257. (1966b). Structure des relations de personne dans le verbe. In Benveniste, Problemes de linguistique generate. Vol. 1. Paris: Gallimard, 225-237. Bryant, Dorothy (1971). The kin of Ata are waiting for you. New York: Random House. Butler, Judith (1990). Gender trouble: Feminism and the subversion of identity. London: Routledge. Cameron, Deborah (1985). Feminism and linguistic theory. New York: St. Martin's Press. Daly, Mary (1979). Gyn/ecology: The metaethics of radical feminism. London: Women's Press. Dubois, Betty Lou, & Isabel Crouch (1987). Linguistic disruption: He/she, s/he, he or she, heshe. In Joyce Penfield (ed.), Women and language in transition. New York: SUNY Press, 23-36. Elgin, Suzette Haden (1984). Native tongue. London: Women's Press. (1987a). The Judas rose. London: Women's Press. (1987b). Women's language and near future science fiction: A reply. Women's Studies Interdisciplinary Journal 14(2): 175-181. (1994). Earthsong. New York: Daw. Fuss, Diana (1989). Essentially speaking: Feminism, nature and difference. London: Routledge. Garreta, Anne (1986). Sphinx. Paris: Grasset. Hofstadter, Douglas (1979). Godel, Escher, Bach. New York: Basic Books. Inoue, Miyako (1994). Gender and linguistic modernization: Historicizing Japanese women's language. In Mary Bucholtz, A. C. Liang, Laurel Sutton, & Caitlin Hines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and Language Conference. Berkeley: Berkeley Women and Language Group, 322—333. Khosroshahi, Fatemeh (1989). Penguins don't care, but women do: A social identity analysis of a Whorfian problem. Language in Society 18(4):505—525. Le Guin, Ursula (1969). The left hand of darkness. St. Albans, England: Granada. (1975). Winter's king. In Le Guin, The wind's twelve quarters. New York: Harper & Row, 93-117. (Original work published 1969) (1976). Introduction to The left hand of darkness. New York: Ace. (1979). Is gender necessary? In Lc Guin, The language of the night: Essays on fantasy and science fiction. New York: Putnam's, 161 — 171.

"She Sired Six Children "

347

(1985). Always coming home. New York: Harper & Row. (1989). Is gender necessary? redux. In Le Guin, Dancing at the edge of the world: Thoughts of words, women, places. New York: Grove, 7-16. Livia, Anna (1994). The riddle of the Sphinx: Creating genderless characters in French. In Mary Bucholtz, A. C. Liang, Laurel Sutton, & Caitlin Hines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and Language Conference. Berkeley: Berkeley Women and Language Group, 421-433. (1995). Pronoun envy: Literary uses of linguistic gender. Ph.D. diss. University of California, Berkeley. (As Anna Livia Julian Brawn.) (1998). Fear of sewers: Who sees this? Who thinks this? Who says this? In Natasha Warner, Jocelyn Ahlers, Leela Bilmes, Monica Oliver, Suzanne Wertheim, & Melinda Chen (eds.), Gender and belief systems: Proceedings of the Fourth Berkeley Women and Language Conference, Berkeley, CA: Berkeley Women and Language Group, 439-446. Newman, Michael (1992). Pronominal disagreements: The stubborn problem of singular epicene antecedents. Language in Society 21(3):447-475. Noel, Georgine (1994). Appelez-moi Gina. Paris: Lattes. Penelope, Julia (1990). Speaking freely: Unlearning the lies of the fathers' tongues. Oxford: Pergamon. Piercy, Marge (1976). Woman on the edge of time. New York: Fawcett Crest. Spender, Dale (1980). Man made language. London: Routledge & Kegan Paul. Spock, Benjamin (1946). The common sense book of baby and child care. New York: Duell, Sloan & Pearce. (1976). Baby and child care. Rev. ed. New York: Pocket Books. Stephens, Inge (1983). Alain, transsexuelle. St. Lambert, Quebec: Heritage. Winterson, Jeanettc (1993). Written on the body. New York: Knopf. Wittig, Monique (1964). L'Opoponax. Paris: Minuit. (1986). The mark of gender. In Nancy Miller (ed.), The poetics of gender. New York: Columbia University Press, 63-73. (1992). The straight mind and other essays. Boston: Beacon.

18

MARY BUCHOLTZ

Purchasing Power The Gender and Class Imaginary on the Shopping Channel

Language and gender in popular culture The vexed question of women's position in popular culture has been answered, broadly speaking, by two opposing schools of thought within feminism. In the predominant, text-based approach, cultural forms—newspapers, magazines, advertisements, films, televisions shows, music—are mined for patriarchal ideologies in structure and content (e.g., Modleski 1991; Mumford 1995; Winship 1980). The utility of textual analyses is confirmed by the continuing resonance of this critical work in women's lives. Nevertheless, such interpretations are brought up short when they expand their scope to include women not only as the objects of popular representation but also as the consumers of them. Given the sexism inherent in popular culture and uncovered in earlier scholarship, women's enthusiastic participation in what is widely understood as "their own oppression" is frequently viewed as deeply problematic. Explanatory theories predicated on women's "self-hate" are not uncommon (Douglas 1994), and analyses often hinge overtly or covertly on the class position of the women who are thought to engage most completely in popular culture. Such theories assume that popular culture is the transmitter of monolithic gender ideologies (see discussion in Strinati 1995) and place the feminist analyst at the opposite pole from the deluded, less-than-middle-class female consumer. In response, scholars who are critical of this outcome have issued counteroffensives from a psychoanalytic perspective that emphasizes personal, even autobiographical, pleasure (e.g., Nochimson 1992). Others have moved away from text-centered interpretation altogether, espousing instead a more ethnographic approach that takes "real" women's cultural analyses as primary (e.g., Christian-Smith 1990; Hobson 1980;Radway 1991). 348

Purchasing Power

349

Both positions, however, suffer from a limited understanding of women's identities and relations to the popular: One views the woman-as-consumer as empowered by her quest for pleasure and the other sees her as mired in oppression. Language and gender researchers have not, for the most part, entered into this debate, although they have a great deal to contribute to it. Whereas textual analysis, influenced by literary criticism, considers popular culture only as a set of "texts" or products to be read and ethnographic studies give primacy of place to the individual consumer, the methods of linguistics, and especially of discourse analysis, allow researchers to examine both aspects and thereby provide a more complete and nuanced picture of women's complex relationships to popular culture (see, e.g., Cameron 1995; Talbot 1995). The fundamental insight of discourse analysis is that "texts," or stretches of discourse, take on meaning only in interaction and that, as consumers of cultural "texts," audiences are active participants in this process of meaning making (Mcllvenny 1996). In particular, as both Jennifer Coates (chapter 6) and Marjorie Orellana (chapter 3) demonstrate in this volume, women and girls do not unthinkingly consume cultural forms but construct their own meanings and identities in relation to such forms. The crucial contribution of this work is the recognition that, because identities are forged in ongoing discourse, they are not fixed for all time but fluid, often slipping beyond the theory-laden explanations of scholars (see also Goodwin, chapter 20, this volume). Discourse analysis also calls attention to the specificities of cultural production: Viewed as a set of discourses, popular culture splinters into multiple and sometimes conflicting representations that offer similarly conflicting resources for women's identity construction. New cultural forms that blend or transcend conventional boundaries of genre or register present special challenges to analysts, for they facilitate new identity formations that may contradict traditional feminist analyses (e.g., Hall 1995) or even traditional analyses of gender itself (Barrett, chapter 16, this volume; Hall 1996). The shopping channel—a pastiche of public and private, entertainment and consumption—is an especially rich example of this discursive instability. As a distinct form of popular culture, home shopping arose with the advent of cable television services and the concomitant expansion of channels from which viewers may select. Although there are or have been several national shopping networks concurrently on the air or in the works that target different markets, such as teenagers, upper-middle-class viewers, and so on, the network that holds the dominant market share (Fabrikant 1994) and from which I take my data is QVC—"Quality, Value, Convenience."1 Because the network is in many ways the archetype of teleshopping, I refer to QVC generically as the shopping channel. The shopping channel is, in effect, a single continuous commercial for a vast array of products. The linguistic practice that results from this genre is a complex type of public discourse that combines the intimacies of private conversation with the exigencies of the marketplace (see also Richardson 1997). Viewers—who are primarily female—are invited into a homelike setting in the television studio in which professional teleshopping hosts, through a variety of talk-based performances, display and describe merchandise for mass consumption. Models may additionally be employed to display such commodities as jewelry, clothing, and exercise equipment. Although the channel airs regularly scheduled "programs" that focus on particular products, such

350

IDENTITY AS IMPROVISATION

as "The Gold Rush" jewelry show, the network's basic format varies little throughout its 24-hour broadcasting day. Viewers participate in this economy by calling the network's toll-free telephone number to purchase products by credit card for home delivery. As the host presents and describes each commodity, the toll-free number is flashed on the screen and the discounted price is displayed while a clock in the corner counts down to the end of the product's availability. The sense of urgency created in this manner is, however, only illusory, because merchandise is made available to consumers repeatedly over the course of several days, weeks, or even longer. If this were the extent of the shopping channel's offerings, it would be difficult to account for its impact on the cultural landscape and its billion-dollar annual revenue. However, the shopping channel differs from other recent and less successful at-home shopping forums, such as infomercials and the Internet, in its emphasis on audience participation, in which home shopping bears a certain resemblance to call-in talk shows.2 This aspect of the shopping-channel experience is nowhere more evident than in the network's use of live, on-the-air telephone interactions between host and shopper. Occasionally, a new caller may be transferred to the host to be welcomed as a first-time teleshopper and to reveal what commodity she has purchased from the operator who took her call. But more often, frequent viewers will call up solely for the purpose of reporting their experiences with a product they purchased through the network. These telephone interactions establish a sense of community between caller and host, and by extension between all viewers and the network as a whole (see also Bucholtz forthcoming). Within this fictive community, callers and hosts discursively negotiate their own and each other's identities in relation to idealized positions of upper-middle-class authority and lower-middle-class authenticity. Callers accrue benefits in terms of both pleasure and power, but at the same time their constructed identities and those of hosts are easily destabilized and reworked within the commercial enterprise of the shopping channel. Station identification: How to be a smart shopper Callers frequently describe themselves with labels that link their identities to commodities: Viewers variously admit to being "a gold fanatic," "a bracelet person," "a jewelry-coholic," or simply a "QVC-ite." But such identifications should not be read as lasting elements of a caller's self-conception. These categories take on their meaning only in the context of consumption-based interaction, and within this frame the caller gains credibility as a discerning and experienced shopper. However, it is only by purchasing products offered by the shopping channel that viewers may gain access to the opportunities for discursive authority that the network may provide. Achieving the privileged position of on-the-air interlocutor with the teleshopping host requires a long-term commitment to viewing—and buying. "I've been waiting for six years to talk," states one waiting caller, and another reports, "I've always wanted to talk and I've nev—this is the first time I've ever been able to talk on the air." The pleasure of discussing shopping on television transcends even teleshopping itself; sentiments like "I'm just thrilled to death" or "This is so much fun!" are very common. These pleasures, of course, are profoundly gendered.

Purchasing Power

351

Women are the channel's target audience, as evidenced by the demographic profile of viewers and callers, by the kind of merchandise offered for sale, and by the discursive practices of the program hosts. Hosts speak directly to the camera, addressing the audience as singular you and gendering this virtual interlocutor as female, as shown in (I): 3

(1)

H

This kind of necklace reminds me of something you might wear to like a jazz concert,

The orientation to women prevails regardless of the gender of the host or the gender associations of the product being promoted for sale: Hosts often suggest husbands as potential recipients of gifts purchased from the shopping channel but rarely gender the viewer as a man who might buy an item for himself or as a gift for his wife.4 The primacy of women in the shopping channel offers them an opportunity for discursive power that may not be available to them in a less feminized realm.5 This privileged status is conferred on experienced teleshoppers, who call in to share their expertise with other viewers. Such callers speak with the voice of authority, which pervades both the content and the form of their discourse, as seen in (2). (2)

(H = host, C = caller) 1 C .hi was calling about the um gold mesh watch? 2 [I believe]= 3 H [Uh huh. ] 4 C =it's: (.) twenty six eighty seven, 5 I'm: not sure. / 6 H /Yeah./ 7 C II think that's the number. 8 1 have the gold one. 9 C) 10 H Do you. 11 C I have to say I get a lot of compliments 12 I mean these people (.) 13 hook at it twice you know because it's different. 14 H Mhm. 15 C You don't see that in the stores./ 16 H /R:ight. 17 [] 18 C [A:nd ] uh you know I really I [really J= 19 H [] 20 C =enjoy watching it 1 would tell anybody to uh (.) 21 you know to purchase it, 22 I don't think they'll be disappointed.

In this excerpt, the caller draws on numerous linguistic strategies to project her identity as a seasoned teleshopper. The caller's discourse is designed to bolster her au-

352

IDENTITY AS IMPROVISATION

thority by offering arguments in support of her endorsement. After establishing that she owns the item and is therefore qualified to comment on it (line 8), she notes that others covet it (lines 11-13) and that it is difficult to acquire elsewhere (line 15). She concludes with a summary statement about the pleasure she derives from the product (lines 18, 20). The testimonial speech event is characterized by this enumeration of evidence, which is cither volunteered by callers or, if necessary, solicited by hosts. Evidence offered in testimonials tends to fall into several broad categories: besides the channel's by words—quality, value, and convenience—callers attest to products' uniqueness, versatility, and ability to elicit admiration from others. Adept callers manage to incorporate all six of these categories into their testimonials. Associated with the normative content of testimonials is a set of conventional linguistic practices used both by callers and by hosts. For callers, shopping-channel discourse introduces the everyday, private pleasures of shopping itself into a more public, formal domain in which prior shopping experiences are shared as information for the benefit of other shoppers. This specialized knowledge is imparted with an equally specialized vocabulary that lends an aura of science and seriousness to the shopping enterprise. In addition to greater formality, the register is distinguished by its large descriptive and technical lexicon. Thus purchase replaces buy in line 21 above; elsewhere callers predict great enjoyment for those who buy a recommended commodity or praise a product as delicate or exquisite. Such word choices also suggest refinement, a displayed appreciation for the finer things in life, which in turn is linked to the complex class relations of the shopping channel (see the following discussion). As a result of the abundance of descriptors in the shopping-channel lexicon, tremendously dense descriptions are applied to even the most mundane items, thereby producing the heavy noun phrases that typify this register and contribute to its technical tone (example 3): (3)

H

It's the seven and a quarter inch Z link tennis bracelet in Diamonique and go:ld and you're gonna absolutely love it.

The register involves additional syntactic quirks that pragmatically invoke shared knowledge. Thus referring terms are often assigned definite determiners even on first mention, signaling that the referent is salient in the discourse, as illustrated in line 1 in example (2) (the um gold mesh watch), as well as in example (3). Because homeshopping discourse is reiterative, with the same products offered for sale again and again, the channel's entire inventory of merchandise is perpetually available for mention. Thus it is constantly susceptible to definite marking. By virtue of this assumption of shared background, speakers may also indicate the discourse relevance of referents in other ways, such as by deleting the head noun phrase in referring expressions, as in the examples in (4): (4a)

C —>

I bought the sixteen inch Figaro necklace?

H Well, I'm ghad you got the Figaro,

Purchasing Power

353

(4b)

—>

H What are you buying tonight Lisa? C Well, I got the advanced order for the PPinocchio. H Wo:nder[ful ] C [Film.J

H Once again, advanced orders only for our Pinocchio,

In the most extreme invocation of shared knowledge, references to commodities are made without the use of any description at all but rather merely by reciting their catalog numbers, as in Example (5a): (5a)

1 2 3 4 5 6 7 8 9 10 11 12 13

C I'm ordering. a graduation gift. for my niece, who's graduating from college, H Uh huh. C and she's a very (.) occelerated speech rate> {small blonde} petite (.) young lady, H Mhm. C and so I ordered J. two seven. oh seven. six.

The possibility—indeed, the likelihood—of successful reference using a number underscores the fact that membership in the world of the shopping channel is available only to longtime viewers. Any competent participant in the discourse must display this referential facility, a task in which hosts are aided by computer monitors, hidden from viewers, that supply important details about each item as it comes up for sale. Nevertheless, on-the-air telephone interactions can become a test of memory for both callers and hosts, whose conversation may introduce products that do not currently appear on the computer or television screen. In fact, callers who have mastered the shopping-channel discourse may prove themselves to be even more knowledgeable than the experts—that is, the network hosts—as illustrated in example (5b), which continues the exchange begun in example (5a):

(5b) 14 15 16 17 18 19

H Two seven oh seven six. Which one is that./ C /That's the: (.) H [.h ba-the (.) yes! ] C [the bracelet and necklace, ] the three tone?

354

IDENTITY AS IMPROVISATION

20

21

H

Yes.

The one that I have o:n.

Using a string of numbers referentially, the caller manages to stump the network host (lines 10-15). But her deft exhibition of discursive competence does not end here: in response to the host's confusion she provides a gloss of the catalog number, the bracelet and necklace (line 18), which, with its definite determiner in first mention, is itself part of the home-shopping register. She finally supplies another feature of the register, a headless adjectival phrase, to narrow down the item further still: the three tone (line 19). The caller gradually offers additional information, using increasingly specific referring expressions that are characteristic of shopping-channel discourse. In so doing, she prolongs her display of her own competent, insider identity at the expense of the host, who does not recognize the catalog number despite the fact that she is wearing the item under discussion. Such exchanges allow callers to take on interactionally authoritative roles, but this authority is sharply limited and sometimes entirely revoked. The redefinition of callers' discourse is a consequence of how class identities are constructed and projected on the shopping channel.

The shopping channel as a class act The shopping channel's preoccupation with class positions is reflected in the frequency with which callers invoke the concept, as marked by arrows in examples (6) through (8). (Each exchange is between a different caller and host.)

(6) H

C H C —>

Well, I'm gl:ad you got the Figaro, it's nice Decelerated speech rate> {to be able to) have a shorter length when you get into (.) the uh deeper necklines that come in for spring and summer. Yeah. / /M[hm. ] [Yeah.l I like the (.) the look of it.= =It's real classy.

(7)

C H C

—> H C

I wear- wore it on two job interviews, .h with a silver gray (.) pinstripe suit, Oo:. And it's justit's so cla(h)ssy. Did you get the job? (.) No.

Purchasing Power

355

(8)

C H C —>

Th:ankyou:! You're welcome Nancy, it was my pleasure to speak with [you. ] [And-] thank QVC for the class, act.

Callers' approval of class arrangements on the shopping channel, however, is not echoed by the network's outside commentators. A New York Times article headlined "Television Shopping Is Stepping Up in Class" opens with the overt snobbery that one would expect from the highbrow Times: "Gone are the hallmarks of the current crop of [home shopping] shows: the cheesy sets and dowdy hosts and the hours of hard-sell pitches for budget goods like cubic zirconium rings. . . . Home shopping executives have seen the future, and it isn't the fiftyish housewives from Dubuque" (McMurray 1994).6 This excerpt reveals the class dimension of the taste promoted by the shopping channel (cf. Bourdieu 1984), a dimension that is here linked to region, age, and, crucially, gender. However, the journalist, in charging hosts with "dowdiness," confuses the taste that the host professes to have and the class identity that she or he actually displays in self-presentation. This construction of class difference plays out in the network largely through discursive practices of displacement, a strategy for projecting class identities that Stanley Aronowitz (1992) has observed in prime-time television as well. Aronowitz notes that in many television shows working-class identity is displaced onto lower-middle-class characters—for instance, the cop is television's paradigmatic working-class male, although the occupation of police officer is not working class.7 An analogous displacement occurs in the shopping channel, where lower-middle-class taste is mapped onto representative middleclass or even upper-middle-class bodies, those of the network hosts. The purposeful mismatch between ascribed taste and displayed identity obscures the obvious class differences between viewers and the idealized consumer typified by the host by suggesting that middle-class status is associated with—and perhaps is even achieved by— the consumption of advertised products. But the distance between the caller and this ideal cannot be bridged so easily; notwithstanding the premise of an L.A. Law episode that first aired in 1994, frequent teleshoppers are not generally granted the opportunity to become telehosts. In the host, the central figure in this drama of consumption, viewers are presented with a middle-class professional who, unlike most callers, speaks Standard English with little or no trace of a regional accent, yet whose reported tastes and consumer desires are representative of the lower middle class. The anomaly of apparently middle-class speakers selling lower-middle-class goods is, moreover, an arrangement that runs counter to the findings of William Labov' s (1972) groundbreaking study of speech in New York City department stores. Labov suggests that sales clerks are in lower social strata than the clientele they serve; as he comments in support of his class analysis, "C. Wright Mills points out that salesgirls in large department stores tend to borrow prestige from their customers, or at least to make an effort in that direction" (1972:45). In the discourse of the shopping channel, the opposite influence is

356

IDENTITY AS IMPROVISATION

at work: Customers strive for the prestige of the sales representatives (that is, the network hosts). This reversal is partly due to the mediated nature of the shopping channel. Hosts are not simply salespeople but television personalities, not mere facilitators of consumption but role models for consumers. Moreover, department stores and shopping malls increasingly aim at a upper-middle-class clientele, leaving lowermiddle-class shoppers feeling that they and their tastes no longer belong. "As the malls have upscaled, they make it more and more uncomfortable for those of us who haven't," remarks a frequent QVC shopper interviewed by journalist Elizabeth Kaye. Kaye's rather condescending article on the network, published in the tony men's magazine Esquire, simultaneously reports on and reinforces this apprehension:8 '"I wouldn't go to Worth Avenue,' says a QVC shopper for whom the most ample size sold on QVC, a size 3X—equivalent to size 24—tends to be a trifle snug. 'Salespeople don't expect me to spend money and don't rash to my side. I would feel out of place. I would feel ignored'" (Kaye 1994). Yet even while the taste of the lower middle class is celebrated on the shopping channel, its voice is displaced from the discourse, for it has been taken over by the more authoritative voice of the host, just as the model-like body of the host stands in for the unfashionably large body of the typical viewer.9 The displacement of the lower-middle-class voice does not, however, lead to its complete disappearance, because it offers something that the shopping-channel host lacks by definition: authenticity. Viewers' class identifications beyond the shopping channel thus shape their interactional identities within it. New and improved: Language repackaged The language of most callers to the shopping channel indicates that they are culturally, though not necessarily economically, located in the lower middle class. That is, most callers have at least enough disposable income to make regular purchases from the network, but the taste that the channel cultivates is decidedly counter to the elite urban, bicoastal aesthetic. The class identity of shopping-channel viewers, however, cannot be read off from audience demographics, which are, in any case, closely guarded by network executives. Rather, it is a construct that emerges interactionally in shopping-channel discourse and, as already seen, in representations of the network in the wider U.S. culture. Callers' identification with lower-middle-class "Middle America" is manifested in their regional accents and use of nonstandard forms. Examples of these are marked with arrows in (9) and (10): (9) C

—>

Urn:,

I get a lot of gold from youse, and it's beautiful.

(10)

C —»

You know another thing I was gonna sYou know them: uh—

Purchasing Power

357

What's her name that (.) uh exercise [person] [Kelly ] [Moretti ] C [Yeah. ] You know them: uh— you know for your arms ? H Yes. Oh:. C You know I was wondering . . .

H —> —>

Callers to the network construct authoritative identities for themselves on the air, but in the context of the interaction between caller and host what began as a discourse of authority becomes a discourse of authenticity. This reworking is motivated by the fact that expertise is in no short supply on the shopping channel: Hosts are of course well versed in the network register, and their descriptive presentations of merchandise are supplemented by guest appearances by vendors, exercise trainers, celebrities, and other specialists. What the shopping channel lacks, and what viewers' language offers, is authenticity. On a network that sells ersatz diamonds and lookof-leather Western wear, believability is perpetually at risk. The warmth and enthusiasm of hosts can go only so far in fostering the sense that the shopping channel is a world peopled with just plain folks who want to bring viewers quality, value, and convenience. The discourse of the callers, refracted through the lens of class relations, provides this missing element. Because callers' language is marked for class and region, it assumes a very different value in juxtaposition to the standard, middle-class speech of network hosts. As has been shown in numerous subjective-reaction and matched-guise tests, a middle-class voice carries authority or prestige, but it may lack authenticity or trustworthiness to listeners of lower social classes (for a survey of this research, see Fasold 1984). For this reason, on-the-air telephone conversations with viewers are vital to the network's marketing success. The following examples illustrate how the linguistic practices of the host come to participate in a discourse of authority and expertise and how the practices of the caller are relegated to a discourse of authenticity and eyewitness testimony. This role assignment is due not merely to the juxtaposition of the two interactants' language varieties but to the discursive efforts of the host. In (11.), for example, the host controls the direction of the discourse, asking the caller to frame her relation to the commodity in terms of pleasurable experiences associated with it (lines 4, 7, 11, 14, 21-26). The host reshapes these reports into a general analysis of the product's quality (lines 29-35). What is paramount in the discourse of the caller, as recast by the host, is authentic experience with commodities and the pleasurable feelings that this experience engenders. The host's discourse, conversely, is characterized by the control of numerous facts about the item. (H) 1 2 3

C

Um I have K one zero seven nine? The French white nine-piece? H Mhm. (.)

358

IDENTITY AS IMPROVISATION

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

How do you like it? C Oh I love it. I use it for everything. H That's a nice set isn't it? C Yes. It is. (.) H And uh what is: the thing that you enjoy cooking in it most? C [Uh] H [Or ] which of the pieces do you enjoy the most? C Well, 11 use all of them, h Urn. I do chicken a jot ancj um j \fac us j n g (.) {that for that.}/ H /Mhm. And(.) do you find um (.) that if you're using something if you're doing like chicken on: uh an- any of the pieces like maybe the the two and a half quart that oval one? It- everything cooks nice and evenly in it too? C Yes. And it seems like it doesn't take as long. H Right. You know, Corning is: has been around forever. I mean if you (.) have had any Corning pieces ever in your life you probably still have them. Because they're so durable.

At the end of the exchange from which this excerpt is taken, the host sums up the call for the viewing audience, casting it entirely in terms of the caller's emotional response to her ovenware: "Liz was dialing in to tell us she absolutely loves her French white ovenware set," he reports. Although the caller in (11) seems to acquiesce in the consignment of her discourse to the realm of testimonial, in example (12) the caller actively attempts to position herself as an expert. However, her voice quality undermines this project: her hoarseness throughout the interaction detracts from the authority of her assertion that her ionizer was effective in treating her allergies. Yet the hoarseness undoubtedly lends a good deal of authenticity. The host by contrast maintains a convincing authoritative stance, immediately launching into a lengthy scientific discourse (lines 35—66) in response to a question from the caller.

Purchasing Power (12)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

H Hello you're talking to us live on the air, Who am I speaking with? C

/ {You can wear it with anything!} / /Mhm. That's right.

Purchasing Power

363

In other instances, callers' testimonials may be cast not as general statements of truth but as personal revelations marked by first-person pronouns and simple past- or present-tense verb forms. Such statements are met with emotional responses from hosts, as exemplified in (16) through (19), which are extracted from earlier examples: (16)

=(4a) C I bought the sixteen inch Figaro necklace?

H —> (17)

-» (18)

—>

Well, I'm gl:ad you got the Figaro.

=(4b) H What are you buying tonight Lisa? C Well, T got the advanced order for the PPinocchio. H Wo:nderful

=(12), lines 29-31 C I'm able to sleep at night, to breathe at night, / H /That's wonderful.

(19) = (12), lines 70-74 H So you really have noted uh a big difference? C Oh indeed. [Yes. Uh huh.] -> H [Great. Great.] —> Well I'm glad we could help you out with it, —> and I hope it continues to work for you.

Finally, hosts may evaluate callers not only on informational or emotional terms but also on the basis of their character and competence, as in line 18 of example (13) (Ruth you are a very very sweet person) or in (20): (20)

C —»

H

It was great to talk to you. It's the first time I got to talk! You did a great job.

It is at this more personal level that hosts themselves become vulnerable to evaluations by callers. Such evaluations comment on hosts' communicative competence in performing their duties on the air by discursively locating them within broader social categories, such as gender and race (examples 21 and 22): (21)

C

{It's a pleasure} to talk to you, you're my very, farvorite.

364

IDENTITY AS IMPROVISATION

H C

H

Thank you. [(I appreciate that.) ] [You always do ] such a wonderful job, but you're always just so,h pre:tty, and (.) always so upbeat, Oh well thank you.

(22)

C H C

H C

H

Uh I just wanted to tell before I start off with what I wantwhat I called [for ] [Mhm] is that you have such a pleasant— I'm sure you've been told this before— a pleasant speaking voice, especially for this hour of the morlning h J [Oh thank you.] And you're very articulate. you know? When you present something?/ /Oh thank you.

In (21) the caller describes the host's appearance and behavior in gendered terms (you're always just so- .hpre:tty, and (.) always so upbeat), while in (22) the caller focuses on the host's linguistic abilities: his "/?/easant speaking voice" and the fact that he is "very arriculate." Given that the host is one of the few African Americans on the shopping channel, the caller's evaluation conjures up a larger cultural stereotype about the speech of African Americans.10 In praising the host's "articulate" speech the caller invokes European American assumptions of African American speech as nonstandard and "inarticulate." Such examples demonstrate that although the hosts act as physical and linguistic exemplars of the middle class, their lower-middle-class audience feels qualified to evaluate their performance of this identity along lines of gender and race. Callers' approving comments about hosts therefore serve as a second kind of testimonial; in fact, the shopping channel explicitly encourages the commodification of those who pitch its products. When QVC shoppers phone in their orders, operators poll them about their host preferences, and hosts are awarded bonuses on the basis not of sales, as on the Home Shopping Network, but of their popularity among viewers. In light of this arrangement, callers may ultimately have greater power to evaluate the language of hosts than the reverse. Conclusion The complexities of language and identity on the shopping channel point up the limitations of traditional feminist approaches to popular culture. Textual analyses are

Purchasing Power

365

likely to interpret the shopping channel as monologic and to discount the active role that shoppers assume in on-the-air telephone conversations. Feminist theories that center on pleasure, however, may fail to see that callers' construction of interactional authority is not unfettered, for callers do not control the means of linguistic production. Thus women's discourse in mass culture faces far more constraints than exist in alternative cultural forms such as zines (see Sutton, chapter 8, this volume; cf. also Cotter, chapter 19, this volume). Even feminist discourse analysis may suffer from certain blindnesses, for by taking the interaction of host and caller as the central problematic it may render the larger corporate structure invisible. Monotonic research methods tend to yield monotonic—and sometimes monotonous—results. Having cleared the ground with numerous well-focused studies of language, gender, and identity, feminist linguists may be ready to recognize the virtues of "quick-cut camerawork"—analyses that bring multiple aspects of the social world into focus in rapid succession, cutting away before the analyst can make a final pronouncement on their meaning. Particularly in the investigation of complex discursive matrixes such as the shopping channel, such a healthy unsettling of analytic focus may bring into relief the numerous ways that facets of identity combine and recombine in late modern culture. NOTES

I am grateful to John Bucholtz, Colleen Cotter, Kathryn Galyon, Jennifer Gurley, Kira Hall, Robin Lakoff, Anita Liang, and Laurel Sutton for their invaluable help during the researching and writing of this chapter. 1. Among these are several other QVC channels, including a health network, a channel aimed at a younger and more affluent audience (onQ), and the Fashion Channel, which evolved into Q2 and then collapsed back into QVC's main program; Penney's Shop Television Network (later bought by QVC); Montgomery Ward's Valuevision; Spiegel and Time Warner's Catalogue 1; Fingerhut' s "S" The Shopping Network; Black Entertainment Television's BET Shop; Time Warner's Full Service Network; TWA's Travel Channel; Microsoft and TCI's computer channel; and shopping channels associated with MTV, Macy's, and Nordstrom. Home shopping has also gone international in recent years, mainly through joint ventures with QVC itself. Shopping networks have been proposed or have aired in Canada, England, Germany, Japan, Mexico, Norway, and Spain, among other places. Although many similarities exist between QVC and its main competitor, the Home Shopping Network, the latter is aimed at a working-class audience and engages in hard-sell techniques. QVC, by contrast, employs more conversational and information-oriented discourse forms. As I will argue, QVC's audience appeal is largely due to its reliance on this format. 2. Another limitation of online shopping is that women, who are the primary targets of such services, are less likely than men to use computers (White 1994). Such considerations may have been an obstacle to the unveiling of QVC's computer shopping service, Q Interactive. 3. All names in transcripts are pseudonyms. Transcription conventions are as follows: end of intonation unit; falling intonation ,

end of intonation unit; fall-rise intonation

?

end of intonation unit; rising intonation

366

IDENTITY AS IMPROVISATION !

end of intonation unit; high intonation self-interruption; break in the word, sound abruptly cut off

—

self-interruption; break in the intonation unit, sound abruptly cut off

:

length

underline

emphatic stress or increased amplitude

(.)

pause of 0.5 seconds or less

h

exhalation (e.g., laughter, sigh); each token marks one pulse

.h

inhalation

transcriber comment; nonvocal noise

{ )

stretch of discourse over which a transcriber comment applies

phonetic transcription

[ ]

overlap beginning and end

/

latching (no pause between speaker turns)

=

no pause between intonation units

—>

text under discussion

4. Despite the relative dearth of men on the shopping channel (marketing researchers estimate that no more than 30 percent of viewers are male, and the figures are even lower for callers and hosts), they play a central role in the network's discourse, as I discuss elsewhere (Bucholtz forthcoming). 5. It has been suggested to me that this authoritative discourse is particularly prized because it may be rare in viewers' lives outside of the shopping channel, but I am extremely reluctant to make this claim in the absence of any supporting evidence. Indeed, it seems likely that callers' authoritative identities emerge in shopping-channel discourse not because such identities are lacking elsewhere in their lives but because they are already developed and available for use in a new discursive realm. 6. The irony of this assertion is evident in hindsight: Following CEO Barry Diller's highly publicized expansion of QVC, profits dropped sharply. The network has since returned to its roots among middle-aged, lower-middle-class suburban women. 7. See Bonnie McElhinny (1995) for a discussion of the historical and social processes that have led to this shift in class position. 8. That an article on teleshopping would appear in such a venue may seem odd but is in fact entirely consistent with the assiduous efforts of class-conscious publications to differentiate their readership from the lower middle class. Editors of the New York Times in particular seem to be fascinated by QVC: In addition to do/ens of news articles, the Times has run at least three teleshopping-related features in its Sunday magazine over the past 5 years. 9. Kaye (1994) reports that the most popular women's clothing sizes on QVC are 18 and 20. 10. On the Home Shopping Network, only three of thirty-seven hosts are African American (Hayes 1995); QVC's figures may be slightly higher. REFERENCES

Aronowitz, Stanley (1992). The politics of identity: Class, culture, social movements. New York: Routledge. Bourdieu, Pierre (1984). Distinction: A social critique of the judgment of taste. Cambridge, MA: Harvard University Press.

Purchasing Power

367

Bucholtz, Mary (1993). The mixed discourse genre as a social resource for participants. In Joshua S. Guenter, Barbara A. Kaiser, & Cheryl C. Zoll (eds.), Proceeding* of the Nineteenth Annual Meeting of the Berkeley Linguistics Society. Berkeley, CA: Berkeley Linguistics Society, 40-51. (1996). Black feminist theory and African American women's linguistic practice. In Victoria Bergvall, Janet Bing, & Alice Freed (eds.), Language and gender research: Rethinking theory and practice. London: Longman, 267-290. (forthcoming). "Thanks for stopping by": Gender and virtual community in American shopby-television discourse. In Mary Talbol & Maggie Morgan (eds.), "All the world and her husband": Women in twentieth century consumer culture. London: Cassell. Cameron, Deborah (1995). The new Pygmalion: Verbal hygiene for women. In Cameron, Verbal hygiene, London: Routledge, 166-211. Christian-Smith, Linda K. (1990). Becoming a woman through romance. New York: Routledge. Douglas, Susan J. (1994). Where the girls are: Growing up female with the mass media. New York: Random House. Fabrikant, Geraldine (1994). "Don't touch that dial" deal: QVC founders see future in shopping. New York Times (August 13): 17. Fasold, Ralph (1984). The sociolinguistics of society. Oxford: Blackwell. Hall, Kira (1995). Lip service on the fantasy lines. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 183— 216. (1996). Cyberfeminism. In Susan C. Herring (ed.), Computer-mediated communication: Linguistic, social, and cross-cultural perspectives. Amsterdam: John Benjamins, 147— 170. Hayes, Cassandra (1995). Cashing in on the home shopping boom. Black Enterprise (February): 120. Hobson, Dorothy (1980). Housewives and the mass media. In Stuart Hall, Dorothy Hobson, Andrew Lowe, & Paul Willis (eds.), Culture, media, language. London: Unwin Hyman. 105114. Kaye, Elizabeth (1994). The new phone sex. Esquire (May):76. Labov, William (1972). The social stratification of (r) in New York City department stores. In Labov, Sociolinguistic patterns. Philadelphia: University of Pennsylvania Press, 43-69. McElhinny, Bonnie (1995). Challenging hegemonic masculinities: Female and male police officers handling domestic violence. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 217-243. Mcllvenny, Paul (1996). Heckling and Hyde Park: Verbal audience participation in popular public discourse. Language in Society 25(1):27-60. McMurray, Scott (1994). Television shopping is stepping up in class. New York Times (March 6):5. Modleski, Tania (1991). Feminism without women: Culture and criticism in a "postfeminist" age. New York: Routledge. Mumford, Laura Stempel (1995). Love and ideology in the afternoon: Soap opera, women, and television genre. Bloomington: University of Indiana Press. Nochimson, Martha (1992). No end to her: Soap opera and the female subject. Berkeley: University of California Press. Radway, Janice (1991). Reading the romance: Women, patriarchy, and popular culture. Chapel Hill: University of North Carolina Press. Richardson, Kay (1997). Twenty-first-century commerce: The case of QVC. Text 17(2):199223. Strinati, Dominic (1995). An introduction to theories of popular culture. London: Routledge.

368

IDENTITY AS IMPROVISATION

Talbot, Mary (1995). A synthetic sisterhood: False friends in a teenage magazine. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 143-165. White, George (1994). On-line mice aren't stirring: Cyber shopping lacks appeal of trip to mall. Los Angeles Times (Dec. 15):D1. Winship, Janice (1980). Sexuality for sale. In Stuart Hall, Dorothy Hobson, Andrew Lowe, & Paul Willis, (eds.), Culture, media, language. London: Unwin Hyman, 217-223.

19

COLLEEN COTTER

From Folklore to "News at 6" Maintaining Language and Refraining Identity through the Media

he linguistic subdisciplines of language and gender on one hand and language preservation on the other have rarely overlapped despite their iiiii shared concern with the relationship between language and identity. The .S-. traditional lack of interaction between the two fields is attributable to biases on both sides: language and gender scholars overwhelmingly study speakers of English and other thriving Western languages, and researchers of endangered languages are generally more focused on cultural and linguistic identity than on gender identity. Yet the revitalization of a minority language that is on the brink of extinction provides a rich opportunity to consider gender and other aspects of identity in a particular situation of use, and gender-based analyses may suggest promising new directions for language-maintenance programs. The approach in this chapter involves a blend of ethnographically and interactionally oriented examinations of language and culture, the language being both the Irish language and the discourse forms specific to media, and the culture being the symbolic referents of the community within Ireland, as well as the practices and norms of the discourse community of media practitioners. From the turn-of-the-century emphasis on collecting folklore to the turn-of-themillennium focus on integrating the Irish language in "modernizing" domains such as the news media, proponents of Irish language and identity have long appropriated textual strategies for renewing what are considered to be irreplaceable resources. In the process, women's roles in the public domain—if not entirely their voices—have expanded, so that their presence and fluency with media forms currently reflect their status as full participants in a prestige sphere. The analysis of the news data under consideration here, taken from Irish-language radio, is an attempt to illustrate that competence in the use of media-specific discourse forms accords women (and men) 369

370

IDENTITY AS IMPROVISATION

a functionally neutral linguistic snace in which gender identities are set aside and a primary identity as media practitioner is negotiated. The data comprise an interview on the Bosnian situation that aired on a currentaffairs program broadcast in the Dublin area in the summer of 1995. In the course of my analysis, T refer to the enfolding contexts to which the interview is linked: Irishlanguage radio, the larger frame of media, its operation within a preservationist paradigm, the position of language as a symbol of nationalism, and gender in relation to the participant roles that can be occupied in Irish life. This study thus filters the data through the larger issues of media language, language revitalization, and the effectively degendered forms of a high-status public discourse domain through a detailed analysis of a single speech situation. The Irish case as a whole also provides a source of fruitful investigation across disciplinary lines and indeed argues for the necessity of interdisciplinary considerations. It allows us to consider the place of folklore genres and other conventionalized discourse forms in the construction of a national ideology and the feminist reflection on the omission of women and women's stories in this construction. It provides a counterexample to the received assumptions of the role of media in the worldwide decline of minority or lesser-used languages. It demonstrates the utility of discourse-level research on obsolescing languages. It allows a new look at women's participation in the dynamics of language maintenance. And, importantly, it reinforces the growing body of research that argues for the variable significance of gender identity, while recognizing the role of the larger social context in the articulation of gender and other parts of the self.

The Irish language in Ireland Ireland is a mostly English-speaking country that nonetheless reveres its heritage language, Irish.1 In Ireland, as in many once-colonized bilingual societies throughout the world, the heritage language remains a symbol of national identity but has relatively few fluent speakers (10,000 to 25,000 Irish citizens report fluency, with fewer than 3 percent speaking Irish on a daily basis, according to recent surveys; Irish Times, March 22, 1994, January 15, 1999; 6 Murchu 1985; 6 Riagain 1992). Although people in principle support strengthening the use of Irish in public and in powerful contexts such as media and government, in practice no one wants to give up English. For this reason and because of historical factors deriving from past centuries of British colonial domination, the language is in serious decline; some Irish citizens declare it already dead (Hindley 1990). The language is most robust in the Gaeltacht areas on the northern and western fringes of the country, far from Dublin. The Gaeltachts arc the historical strongholds of the language and have long been the focus of preservationists. With respect to language attitudes in the late twentieth century, the ability to use the native language as spoken in the Gaeltachts is seen as highly desirable. This modern orientation toward the language and its speakers, most of whom are bilingual, is overlaid on a past of negative attitudes and "internal colonialism" that afflicted all of Britain's "Celtic fringe" (Dorian 1981:19).2 Negative attitudes

From Folklore to "News at 6"

371

toward the language and its rural, uneducated users persisted into this century. The speakers' low social and economic status was linked to their use of a disfavored language, making it seem only logical to eliminate the language as a first step to eliminating stigmas of other kinds. The undisputed economic advantage of knowing English has hastened the process of language loss in this century, but at the same time Irish has managed to remain strong as a symbol of Irish nationhood. In an act that linked linguistic pride with national pride, language with nationhood, Irish was named the first official language when the Republic of Ireland was formed in 1922. Language-attitude surveys conducted by the Instituid Teangeolaiochta Eireann (Linguistics Institute of Ireland) as recently as 1993 indicate that this symbolicnationalistic link remains strong in the national consciousness. Some 73 percent in the survey agreed that "no real Irish person can be against the revival of Irish." Furthermore, attitudes toward the value of the language have improved in the last twenty years: 31 percent of respondents agreed that Irish is a dead language, down from 42 percent in 1973, and the percentage of respondents who felt that Irish could be revived for everyday use rose from 39 to 45 percent in the same 20-year period (Irish Times, March 22, 1994). Additionally, the results of the 1996 census, published in December 1998, reported an increase in Irish use by preschool children to 4.5 percent as compared to 3 percent of adults (Irish Times, January 15, 1999).

Women and language maintenance Although scholarship on endangered languages rarely highlights the role of women in the preservation of heritage languages, numerous revitalization projects make use of women's traditional cultural roles to promote the use of the traditional language. This tendency may be an extension of women's stereotyped role as conservers of language (e.g., Labov 1990). Several projects build on women's positions as caretakers of children to foster the "mother tongue": In Hawaii, for example, parents (mostly mothers) serve as classroom aides in Hawaiian immersion programs (Hinton 1994a), and in the California-based Master-Apprentice Program led by Leanne Hinton (1994a, b), gender plays a role in how apprentices learn Native California languages and how they pass on their knowledge to younger generations. Earlier in this century, California Indian women who defended their children, their language, and their culture against the destructive forces of assimilation used their authority as mothers to speak out (Dobkins, chapter 9, this volume). And even when women step into nontraditional roles to lead efforts to preserve the heritage language, they may still be constrained by male dominance in the traditional culture (Craig 1992). Although in these cases women assume leadership positions in their capacity as women, in the Irish situation women tend to contribute to language maintenance through the adoption of public, professional roles that do not hinge on their gender identity and that do not rely on stereotyped beliefs about women's greater linguistic conservatism. In particular, women's representative numbers on Irish-language radio and other contemporary broadcast media projects enables them to participate in a language-revitalization project that modernizes the role both of the Irish language and of Irish women in contemporary society.3

372

IDENTITY AS IMPROVISATION

The use of the media to enhance the status of a language Media language is public language, and as such it is a primary source of cultural messages. The language of media functions both as an agent and as a mediator in the construction of social identity. It reflects cultural norms and positions while at the same time offering a site for their contestation. To better understand the power that the media licenses or inhibits, it is well to examine closely the practices and assumptions that underlie the production of media language while situating these practices within their social and historical context. The media have been blamed by linguists for their role in the recent accelerated obsolescence of languages (e.g., Krauss 1992), but in fact how they play out this role has not been explored. Moreover, there are several counterexamples to the generalization that the media contribute to language death: A number of advocates of minority languages use the mass media for their own ends, from resisting deterioration of the heritage language to contesting outright the power relations that the majoritylanguage society imposes (Brody 1995; Jaffe 1994; Spitulnik 1994). Although the concern about the media is warranted (it is estimated that some 80 to 90 percent of the world's languages will decline in the next century), discussions about the media as a social dynamic and their role in the formation of identity have been fairly superficial. The detailed presentation of the example of Irish (see also Cotter 1996a) is intended to contribute to an expanded discussion of the media's role, positive or negative, in language maintenance and the retention of identity. Language workers in Ireland currently use broadcast media (both radio and television) to enhance the status of the Irish language. Irish-language media give a public—and legitimating—voice to a language historically denigrated by Englishspeaking power-holders, and consequently the language is seen as capable of coping with the discourse needs of the modern world. In short, the status of the Irish language is "reinvented" through the media. And in the process, a new role for women is articulated. Radio has been used explicitly for nearly 30 years as a tool for the preservation and growth of the Irish language. (Ireland's example presaged a policy of the European Union's Bureau of Lesser-Used Languages, which endorses incorporating media in any language-preservation program.) The two most prominent Irish-language stations in Ireland, which have worked to extend the language into modern contexts, are Raidio na Gaeltachta, based in the rural, formerly isolated Irish-speaking areas of the Gaeltachts, and the Dublin-based Raidio na Life. Raidio na Gaeltachta Raidio na Gaeltachta (RnaG) got its start in the early 1970s in the wake of nowhistoric protests by language activists in the Western Connemara Gaeltacht. Unhappy with Irish-language tokenism on the national radio and television stations, the activists started an unauthorized "pirate" Irish-language station in Connemara, the most populous of the Gaeltachts. The government responded by officially establishing and funding an all-Irish radio station in 1972. Declan Kiberd recounts the events leading to the formation of RnaG:

From Folklore to "News at 6"

373

In 1969, inspired by the Civil Rights movement for black emancipation in the United States, a group of activists in the Connemara Gaeltacht launched their own campaign to revitalize the Irish-speaking areas. . . . the Cearta Sibhialta (Civil Rights) movement was in most respects remarkably successful.. . . [lit managed to detach Irish from the purgatorial fires of the school classroom and to present it as part of a global countercultural movement constructed upon "small is beautiful" principles. (1996:567-568)

Indeed, as the only station in the world broadcasting to ethnic minorities at the time, RnaG was a "trail-blazing service in '72," according to Padraig 6 Duibhir, manager of broadcasting services at Raidio na Gaeltachta. Now throughout the European Union there are some thirty to forty radio stations broadcasting in the minority languages of Europe. And RnaG, with its professional mix of news and currentaffairs programming, has become the standard-bearer of the language and a model for quality Irish-language radio broadcast practice. Women, as well as men, are significant contributors to this practice in all areas of station work.

Raidio na Life Aiming to fill the perceived gap left by RnaG, the programmers of Raidio na Life (RnaL), which has been on the air since September 1993, pitch their programming to the urban dweller, especially targeting young Irish speakers who do not have the benefit of a Gaeltacht to promote the linguistic solidarity and exposure that RnaG achieves. (Whereas RnaG is transmitted countrywide, RnaL's community license signal extends only 18 miles into the greater Dublin area.) The station utilizes speakers on the air whose Irish varies greatly in fluency, as well as those who speak a variety, Dublin Irish, that is low in prestige but nonetheless is spoken extensively by the bilingual natives of the urban area. RnaL also deliberately appeals to a younger audience by using contemporary music instead of only traditional music—an equally revolutionary decision within the Irish preservation context. According to Eamonn O Donaill, director of the Irish Language Centre at University College Dublin: It is arrogant lo say thai young people ought to listen to the kind of music that interests us and that they are not properly Irish unless they do the same! This sort of thing puts people against the language. Irish isn't a package deal—a person should be able to be interested in the language but ignore other aspects of the culture if he or she is inclined to do that. (6 Donaill 1995; translation mine)

While 100 percent of the spoken word on RnaL is Irish, 80 percent of the music programming, drawn from all over the world, is in English. By offering mainstream and world music, it is RnaL's intention to attract attention to the language via popular culture and discussion of contemporary concerns. According to O Donaill in a 1996 interview for this study, Irish is associated with tradition in the young person's mind: "The majority of young people think Irish is unsophisticated, and they have difficulty expressing ideas through Irish. About [the traditional Irish practice of] cutting turf they have no difficulty . . . but mass culture and TV particularly, they

374

IDENTITY AS IMPROVISATION

have great difficulty and they switch to English for that" (O Donaill, personal communication). The effects of Irish-language radio on maintenance and identity The two stations have accomplished a great deal with regard to language revitalization. RnaG's stated goals are to use language to establish links in the Gaeltacht community and to enhance the contemporary status of Irish. A logical third goal, which is not strictly part of RnaG's ideology, would be to alleviate a historically ingrained linguistic insecurity. RnaL more explicitly addresses this issue. RnaL's goals are to evoke language change from within. Specifically, the station intends to give people an outlet for the language they learned in school and a reason to learn it; to provide a service to urban Irish speakers who would otherwise be isolated from one another, particularly those attending all-Irish schools; and to show people that the language can adjust to modern life, that one can talk about anything through the medium of Irish. RnaG appears to have been successful in fulfilling its intentions to establish inter-Gaeltacht connections. RnaG both differentiates the dialect areas with its explicit broadcasts from the three regions and unites them through the common language and through the temporal and discourse structures of the medium. Unity among the Gaeltachts is also achieved through the content of the local broadcasts by reinforcing knowledge of community patterns, practices, and values, including artistic ones, which are held in common across the dialects since they all share a similar socioeconomic history. The result is a sense of the importance of one's own dialect and its connection to the language overall. The structures of the broadcast genre, borrowed as they are from the Englishlanguage broadcast milieu, somewhat paradoxically reinforce a feature of the dominant English-language status quo for the purposes of linguistic continuity and vitality. This practice can be viewed as a necessary maneuver for garnering credibility, particularly when the minority language has little inherent or historical status in the public or institutional sphere, as is the case with Irish.4 Simultaneously, the local content programming on RnaG reinforces a sense of Gaeltacht community. Radio programs cover items of Gaeltacht interest and promote traditional verbal and musical art forms (which, according to Watson [1989], the English-language media have helped to erode), as well as hybrid musical forms that rely on an interplay of traditional and contemporary resources. The combination of Anglo discourse structure and Irish-interest content makes the language seem both normal (by using familiar structures of the dominant language community) and special (by using the referents of a once-stigmatized politically powerless community in the public, legitimizing sphere). The nature of this blend, that is to say, the negotiation of dominant ideology with the "core values" (cf. Smolicz 1992) or linguistic attributes of the minority-language community as expressed through the language of the media, remains largely unexplored. The activities of this "intertextual gap" (Brody 1995) appear to vary according to social and

From Folklore to "News at 6"

375

linguistic history, offering evidence to support various theoretical positions about language change (e.g., Thomason & Kaufman 1988) or to answer questions of social meaning posed by language-obsolescence researchers such as Kathryn Woolard (1989). With two decades of RnaG paving the way, RnaL is able to take another direction, filling certain gaps, particularly with respect to addressing the rampant linguistic insecurity in speakers who are not from the Gaeltacht. On the local level—in the RnaL studio itself—volunteer staffers who start out with only a little School Irish speak the language with confidence a few months later. Everyone is given a job and made to feel that her or his contribution is important. This extends to the linguistic realm. In the small world of the RnaL studio, the community of speakers has managed to alleviate linguistic insecurity and make the use of Irish (and a disfavored Dublin form at that) a high-status endeavor. Because RnaL historically has been more interested in affording young speakers an opportunity for practice than in meeting traditional standards of usage, the language on the air is often marked by forms that accomplish a speaker's communicative goals at the expense of idiomatic Irish. Such disregard for the niceties of linguistic prescriptivism is unusual in the media: Indeed, in many countries the broadcast variety is synonymous with the standard (Milroy & Milroy 1991). The needs of both audience and radio workers are considered in RnaL's approach, which focuses on building an Irish-speaking speech community in an urban context not historically sympathetic to the cultivation of such a community. The distinctive feature of mass communication—the disjunction of place between community and audience—actually works in RnaL's favor. A nontraditional location, the airwaves, is used to create a nontraditional community (cf. Bucholtz, chapter 18, this volume, and Dorgan 1993 for other examples of how the media facilitate the formation of nontraditional communities). For speakers lacking access to and cultural affinity for the traditional rural strongholds of the language, RnaL creates its own place, in essence its own Gaeltacht.5 One of the factors that has stood in the way of a reemergence of Irish is that it has been "deprived of contemporary status," according to the Gaeltacht Authority's Padraig 6 hAolain (personal communication). His view is shared by both RnaG's and RnaL's personnel. A slogan of the Gaeltacht Authority, which promotes the economic, as well as the linguistic, interests of the various Gaeltachts, is "Normalise to popularise." If Irish citizens see and hear Irish being spoken on the radio and television, the reasoning goes, its prestige rises and its former "rural" taint disappears. What is normal becomes popular, no longer stigmatized. Likewise, the "modernization" of women's roles through their equal participation in the broadcast media could very well work to make such gender equity in the public sphere a normal part of Irish life. It could also help to alleviate the resistance to women's participation in the public sphere that some people have observed both historically (cf. Foley 1997) and in the modern day. Gaffney, for instance, notes that 50 percent of Irish people now believe in gender equity at home and work and suggests that "what is required is to actually begin to act, even in a small way" (1996:185). Although it is not the explicit goal of either station to integrate women into public life, their efforts to integrate Irish into public life have offered women easier access to participation in public discourse because the media create positions that must

376

IDENTITY AS IMPROVISATION

be filled.6 The stations' underlying methods are complementary. RnaG adopts Anglo discourse forms to give authority to Irish-language media but maintains a strict policy of Irish-only content, preserving the language as it is spoken in the Gaeltachts, the traditional repositories of Irish.7 It thus values the prestige language forms traditionally associated with "core" Irish identity. RnaL, on the other hand, innovates with a hybrid form that challenges assumptions of the language's position in Irish life, as well as the structural form of the language itself, and meets the needs of urban speakers, creating a "Gaeltacht of the Air," as I term it. In so doing it provides opportunities for urban professional women to take up authoritative roles in the media even in the absence of fluency in standard Irish (and several announcers, female and male, have gone on to careers at RnaG and at the Irish-language television station, Teilifis na Gaeilge). Both stations facilitate Irish-language competence among interlocutors in their respective speech communities, reflecting the linguistic profiles of their target audiences. Together, they have the potential to expand the base of Irish speakers, female and male, in public and private domains, and they make a powerful statement about the social contexts of language use, their positions at either end of the preservation-growth spectrum offering a natural laboratory in which to consider the sociolinguistic and discourse parameters that characterize language and identity in flux. Retraining roles through media language Unlike other language-maintenance situations, traditional gender roles are not an obvious factor in current Irish-language broadcast practice. The ratio of female to male announcers and producers is similar to that anywhere else in the First World (that is, not equal but in greater numbers than in the decades before the women's movement of the 1970s; cf. Mills 1988).8 The steering committee for the new Irishlanguage television station seats three women, and the "female perspective" is explicitly sought out and valued, according to Brian MacAonghusa, who oversaw the start-up of the station (personal communication). This inclusiveness suggests that in the last decade of this century, women have a chance for an equal opportunity to modulate the public voice of the media. Previously relegated to private roles in the home, women are becoming agents in the public construction of a social and cultural identity. This is not a trivial accomplishment. As historian Timothy Foley recounts, although the role of women has always been debated in Irish society— especially in the nineteenth century, of which he writes—the traditional position has always been very clear: "Rocking the cradle and ruling the world were two profoundly unconnected activities" (1997:24). Women were expected to stay within the domestic circle, to maintain the traditions and "moral health of society," and essentialist arguments, in which women's "natures" excluded them from employment and participation in the public sphere, held sway until attacks by feminists of the late nineteenth century confronted the propositions themselves. But this cultural shift raises other questions: To what extent does the current preservation ideology in Ireland ("Normalise to popularise," especially through the media) promote women's voices and women's place in the public sphere? When the

From Folklore to "News at 6"

377

modern preservation movement started in 1894, the collection of folklore was a focal point in the building of a new national identity, which could be articulated through narratives and other linguistic practices. Stories representative of men's genres were collected and framed as the legitimate definitions of culture, the storied equivalents of "News at 6." Genres specific to women—cradle songs, keening, fables told to children—were not (Joan Keefe, personal communication, citing Angela Bourke). Even the stories of the legendary Peig Sayers, collected alongside those of her male neighbors in the Blasket Islands, for a time in the 1960s and 1970s became an object of ridicule and cliche (cf. Bourke 1997). Women can now participate in the public discourse domain through the media (since the church still precludes any major role), but to what extent do they influence it? Is the absence of gendered media discourse a slighting or undervaluation of women's experience, a necessary accommodation to the normative paradigm (which in the Irish case follows Anglo structures for reasons having to do with the colonial past), or an underlyingly subversive act? The early folklore corpora omitted women's stories, which were not even considered for inclusion because they were "of the home," out of the public sphere (in this volume see Sawin, chapter 12, on early folklorists' devaluation of women's narrative practices and feminist folklorists' response). To what degree do modern media miss or dismiss women's stories (cf. Mills 1988)? It is important to realize that an absence of women's stories and voices is not a consequence of the Irish culture itself nor of Irish language-planning policies, overt or implied. Rather, the pattern of exclusion is much wider and more global in scope (Mills 1988); the preservation paradigm only serves to make these issues more urgent because, as most endangered languages continue to decline, time is running out for scholars interested in female speakers of such languages or in exploring the broader social context in which women participate discursively within a culture. Wide-ranging macro-level questions such as these require a finer analysis on the micro levels of language, particularly in a bilingual context such as is found in Ireland in which identities are fashioned through more than one language. One of the claims I make here is that discourse features appear as a result of the dynamic, evolving act of talk and also serve to constitute the structure and meaning of talk and its participant relations; contrary to the assertions of some researchers (e.g., TroemelPloetz 1992) the gender of participants does not necessarily determine the shape of media discourse. In support of this claim I present a transcript of an extended news interview that aired on RnaL in August 1995. My goal is to show how one feature of the talk exchange, the use of English-language discourse markers, plays a part in the negotiation of turns and the position of participants in relation to each other within the situated activity frame of media language. It also demonstrates that in media discourse, gender identities may be backgrounded in favor of identities rooted in the ongoing interaction (see also Goodwin, chapter 20, this volume, on locally relevant activity-based identities in which gender is not always salient). That the markers are in English evokes the fluid identity boundaries that can be characteristic of bilingual discourse, even among speakers in a public radio broadcast. What follows is the English translation of portions of a transcript of a live, on-air interview on RnaL's regular half-hour current-affairs program, Um Thrathnona (in the evening). The interviewer, "Aine Ui Laoire" (designated in the transcript as

378

IDENTITY AS IMPROVISATION

UT or "Um Thrathnona"), is a 22-year-old Dublin native who plans a career in the broadcast media. She learned her Irish in school, graduating with a bachelor's degree in the language. She conies from a family of non-Irish speakers and is considered fluent. Her interviewee, "Maureen Cullen" (MC), was interviewed over the telephone (which accounts for some unintelligibility on the tape). Informant judgments about her Irish indicate she is a fluent Dublin Irish speaker with a great deal of prescriptively grammatical facility. Her talk thus balances what could be seen as the conflicting demands of several identities: urban dweller, woman, Irish speaker, and expert on Bosnia.9 (1)

UT:

Welcome to Um Thrathnona I'm Aine Ui Laoire with you until 6:30, Today, we have a special program about the situation in Bosnia which is getting worse every day. This is a complicated situation and now we have Maureen Cullen to explain the historical background of that country and the troubles there up 'til now. (change of vocal quality] UT: Welcome, Maureen Guillen MC: IThank you.

In (1), Cullen's expertise in the matter of Bosnia is assumed, by virtue of the fact that she is interviewed, but not made explicit. We do not know why she is interviewed beyond her Irish capacity and general facility with the topic. This is interesting from a media-discourse point of view, for the authority of the speaker is backgrounded; the listening audience is not bound to evaluate the discourse on anything but informational grounds. Quite possibly, she is well known for her work in relation to Bosnia, which would be part of the perceived shared knowledge of the intended audience. Also of interest is the assumption of the opening question in (2), Not everyone understands how the troubles started in Yugoslavia. Can you explain simply? The question acknowledges the presence of a listening audience (for whom the program is directed and who will hear, along with Ui Laoire, the answer returned by Cullen) and its possible confusion over the complex political situation in the former Yugoslavia. From the standpoint of media practice, it is advisable in an interview of this length to start with an "easy" question. "Easy" questions, whose answers call up information readily, serve the purpose of putting the respondent at ease—a necessity that goes beyond journalism when one or both interlocutors have limited proficiency in the language (which is often the case on RnaL, though not in this instance). Easy questions also simultaneously invoke or build a shared knowledge base for all participants, be they interviewer, interviewee, overhearers, or listening audience. The initial stages of interaction thus establish the roles of both participants in this exchange; these roles are tied entirely to the goals of the talk and have little connection to gender or other components of speakers' identities beyond this interaction. This pattern

From Folklore to "News at 6"

379

is also evident in the use of discourse markers, especially the marker y 'know. Before considering y 'know, however, I wish to examine the role of other English discourse markers in the interview. Although the extensive literature on women's use of discourse markers tends not to examine their role in gaining authority, such functions are inescapable in the present data. Example (2) contains an instance (in line 2) of the speaker's use of the marker now in a face-saving repair. (2)

UT: Many people don't understand how the troubles started in Yugoslavia. Can you explain this simply? MC: Em I'll try to do that, It's em, as you say it's a complicated situation uh: as many people know Yugoslavia was a united country until five or six years ago. So that em perhaps people have heard about Marshal Tito 1 who was, uh, y'know. ruling the country for uh years since the Second World War. And still Yugoslavia was em six different nationsuh Serbia, Croatia, Bosnia, Slovenia, Montenegro, 2 an:d um: now. I can't remember what the sixth one [unintell?] [speaks rapidly, with raised pitch and some embarrassed laughter]

Here and elsewhere in the Irish data I collected, now is used in its English form for discourse-level purposes that relate directly to the speaker's linguistic or discursive fluency. This function is by no means gender-specific: I mean was used in a similar fashion by a male interviewee in a brief telephone interview with a male interviewer that aired on RnaG. Besides its adverb status, now can function as a temporal marker in "discourse time," to use Deborah Schiffrin's (1987) term, while / mean marks speakers' evaluations of what they are producing in discourse. Understanding such gradations in discourse meaning is the first step to accounting for their presence within the bilingual domain. In the instances of their appearance in the data, now and I mean are both used specifically in repair situations. The repair itself is of a very particular kind, commensurate with the bilingual context of talk: that of restoring monolingual coherence to a turn marked by a lapse of memory (in the case of now under discussion) or Irish-language fluency (in the case of / mean in the RnaG data). In other words, in making their conversational repairs the speakers opt to use the markers of the language-of-comfort, or first language, English, rather than the markers of the language-of-discourse, Irish. Underlying this is a consciousness of the marginal position the Irish language plays within Irish public life and discourse. The interlocutors themselves know that between them and among their audience English is the dominant language. Thus English markers are a resource for discourse organization and coherence when actual or potential breakdowns occur. Markers become pragmatically relevant to the bilingual environment of talk again in Example (3), where the use of so as a discourse-topicality marker is illustrated.

380

IDENTITY AS IMPROVISATION

The interviewer's use of the marker (line 1) initiates its global possibilities: It essentially draws to a necessary close nearly four previous minutes of uninterrupted interviewee talk. Globally, it isolates a main, summarizing point (the Bosnian Serbs' quest for land as a catalyst for the existing problems), and locally it creates a logical transition for the next question, which it is the interviewer's journalistic task to produce. Here the interviewer's use of a discourse marker signals her interactional identity, and the apparent backchannels produced by MC in response (lines 2 and 3) are better seen as devices for regaining a conversational footing than as markers of collaborative speech, as found in some women's private conversations (Coates 1989). (3)

1

UT:

2

MC: UT: MC: UT: MC:

3 4

So. the land is the thing they are trying to get= / I Well, that's it= l=are they trying for anything lelse= I =/yeah I yeah I =to get lapart from the land I IWell, it's: it's: I

Example (4) marks a point at which the interview becomes more interactive and departs from an idealized Q&A/speaker-hearer format which separates roles and functions. This heightened interactivity is evidenced by shorter responses by the interviewee and increased participation by the interviewer. Here, if anywhere, gender identity is most likely to emerge from the discourse. However, this is not what occurs. (4)

1 2

3 4

MC: It is difficult to come to a solution because: there are extremely strong feelings on all sides, I think / uh y'knowUT: /And there is the blame on one side. MC: Well:, there's: um the different groups say um: that there is blame on two sides but it is clear who started, em, in this case It is the, the Serbs, em, the Bosnian Serbs and originally they had support from Serbia itself which is sort of is a different country. Em, it is they who started the whole thing em:/ UT: /And when did it really start. About? / MC: /uh about /maybe [Nineteen] ninety-one, no= UT: /mm MC: ninety-two, ninety-two, sorry, yeah? three years ago now.

From Folklore to "News at 6"

381

uh, it's four years ago that the war started between Croatia and Serbia. [breath intake] And um: 5 since then they are, uh y'know they are not uh uhuh stopping until they have it all? by the looks of it. And even: the Muslims are trying to fight back. But there is: uh an "international embargo." against giving arms, to both sides, to the Serbs and to the Muslims but. it looks like the Serbs are, they were able, em: em (.) to get arms from the East, probably from Russia? and beforehand they had much stronger arms than the Muslims had? [breath]= UT: [quick intake of breath] 6 MC: =y'know? [reclaiming floor] and so the Muslims weren't able to defend themselves [breath] UT: /[latching] In Example (4), the increase in interaction between interviewer and interviewee does not produce an increase in "gendered discourse." UT's questions are aimed at eliciting information, and MC's apparent request for confirmation in line 6 (y'know) actually serves to reclaim the floor—that is, to fulfill her discourse role as interviewee. Example (5), which continues where (4) leaves off, is the closing of the speech event, marked by turn negotiation, floor-holding by the interviewee, overlap, and multiple English discourse markers heretofore unused by the interviewer (line 5). Again, these markers are used in the service not of social identities but of interactional identities as the interview winds down. (5)

1

2 3 4

UT: And do you think the United Nations are not doing their best about [unintell?] MC: I think there isn't enough:, the people there on the ground say [deleted material] they're not interested really, y'know?. ah: They will make various statements but really they don't care at all 1 think [discusses the Russian, French, and U.S. attitudes] Y'know. France and uh the United States uh have made various efforts but. although they could be stronger I thinkY'know that's the trouble [hhhh] Ithere. mm:

382

IDENTITY AS IMPROVISATION 5 UT:

IQka:y, Mauree:n/ \Well. thank you very much Mauree:n ICullcnl= 6 MC: lyeah I UT: =for being with us on Um Thrathnona Itoday I MC: lyou're welcome I UT: Thank you.

Unlike the other English discourse markers, which appear only once or twice in the interview and may be used by either participant, the marker y'know appears prominently throughout the interview (twenty-six tokens total) but is used only by the interviewee, a pattern that suggests that although the interactional tasks are distributed between participants, they are also distributed according to relative role and status. The interviewer, in the position of the ostensible controller of the discourse, is the dominant participant in terms of the communication structure of the media interview (y'know tends to mark the less powerful person in the discourse; Schiffrin 1987). The lack of visibility (because the interview is by phone) exacerbates the need for the respondent to engage with the interviewer (eleven tokens of y'know appear in sequence 2, the lengthiest sequence), for whom the customary backchannel responses of conversation arc not allowed in the media discourse frame. In fact, as the interaction level increases (especially at the point of turn negotiation in example (4) but also elsewhere in example (4) and in (5)), the overall frequency of the markers increases correspondingly. Additionally, the frequency of y'know in the data relates to the level of response by the interviewer (the ratio is higher when the interviewer participates less). There are only two y 'know tokens in the highly interactive example (5), which has five interviewer turns, and eleven tokens in example (2), with only one interviewer turn. Thus the interviewee's use of discourse markers does not suggest the collaborative egalitarian talk typical of some women in private interactions (Coates 1989). English discourse markers occur in an environment in which both the careful use of Irish and attention to the multiple situated discourse needs specific to the media are cultivated. Some of the factors that condition the use of language in this publicdiscourse arena arc news values, a conventionalized manner of organizing information, the role of the audience, and the technical limitations of the medium (cf. Cotter 1996b, in press). Discourse-marker insertion, then, as illustrated in the media data, becomes a strategy for discourse coherence and the negotiation of identity in a bilingual frame. The identity which is negotiated, however, derives from the context of use, from roles that are required by the task at hand and not by the gender of the participants. The discourse data show us aspects of discourse structure and discourse interaction. They do not confer the possibly expected cues associated with women's communicative style (backchannel responses, hedges, and so on). What are improvised and jointly negotiated are the participants' purposes. The interviewer's purpose is to elicit information, produce well-formed discourse as a journalist and Irish speaker, convey information, and conduct her linguistic business fluently (that is, with no dead air or lapses). The interviewee's purpose is to present information, establish credibility, and ensure that she has been fully understood. Because the speakers are ac-

From Folklore to "News at 6"

383

complishing a task activity, negotiation of status is minimal (see also Goodwin 1993). At the same time, the speakers improvise their Irish identities, relying on nontraditional resources (such as English discourse markers) to project fluency in the nontraditional realm of radio discourse. The fact that as women they themselves are nontraditional participants in such discourse is itself an improvisation of Irishness, though not one that is highlighted in the interaction. It is possible, then, for speakers to degender language and wield it as a tool on behalf of their work—not simply of their profession (as, for example, radio announcer or commentator) but also of their larger goals (language maintenance) and their interactional goals (gaining the floor, closing the exchange, and so on). In noting that gender is not always salient as a theoretical category and that its effects may differ across contexts, this analysis supports Amy Shuman's (1993) and Patricia Sawin's (chapter 12, this volume) critique of the limitations of folklorist and feminist categorizations of genre, as well as the findings of Alice Freed (1996) and Alice Greenwood (1996) on the variable relevance of gender in discourse. Although it may be argued that radio discourse is a historically male-constructed form, women in mainstream media generally do not challenge the origination of the tool, and the women on RnaL are no exception, for that has not been identified as the problem to solve in this context. Instead, they are overtly challenging assumptions behind the preservation paradigm, which has been identified by RnaL as the central issue. In other words, neither the gendering nor the degendering of radio discourse should be seen as universal: Instead, these processes are particular to the social or ideological needs of using radio (Leitner 1980). Although women's identities as women do not play a significant role in the Irish data presented here, gender has been found to be directly relevant to the workings of radio discourse in other contexts (Bucholtz 1996; Castillo 1994). The place of gender and other aspects of identity in interaction must therefore be determined on a case-by-case basis by examining the details of discourse. It is also important to point out with respect to women in the media in Ireland that they are occupying roles that were once unavailable to them and in so doing are breaking the boundaries of their socially assigned niches. It is possible that expanding the domain of roles, much as the revitalization movement is expanding the domain of use for language, will create different opportunities and ultimately different judgments about positions women can occupy in the public sphere. As it stands, given the current visibility of Irish-language media and their practitioners, the language is refrained as useful and viable, and women are reframed as competent in the sphere of public discourse. Additionally, this perspective on language and identity as it pertains to gender supports current cultural research into Irish life and ideology by historians and sociologists (cf. Kelleher & Murphy 1997), who see women's and men's roles as more interdependent than traditional accounts would have them. Conclusion The position of the media as multiply modifying, supporting, and contesting the community values and characteristics that mark social identity has been illustrated through the example of Irish-language broadcast media, which enhances the status

384

IDENTITY AS IMPROVISATION

of a language that remains a strong symbol of Irish identity. It also points to the notion of multiple marginalities—the positioning of women in Irish-language radio on the periphery of mainstream academic and political concerns. For example, Ireland tends to be overlooked in postcolonial studies (cf. Kiberd 1996), as well as in English feminism. Women's discourse forms are overlooked in traditional studies of folklore (Hollis, Pershing, & Young 1993). Radio tends to be overlooked in studies of media (Cotter 1996a; Pease & Dennis 1995), and media is not considered to the extent it could be in studies of language use (Bell 1991; Cotter 1996a; Sutton, chapter 8, this volume). Nonfluent speakers, such as those on Raidio na Life, are often disregarded in studies of identity (but see Dorian 1991 for a counterexample), and nonstandard speakers tend to be overlooked by preservationists in favor of standard speakers. The Irish case is intended to illustrate generally a complex of points that researchers from different domains can fruitfully pursue and to consider specifically how media—through language and interaction—operate as a dynamic in the evolving process of a society's identity formation: locally, culturally, and nationally. As such, the work here is also intended to remind researchers that gender is not necessarily salient in interactions in the ways students of the subject have come to expect. The salience can be foregrounded, however, only as we consider the norms of the dominant discourse community (Cotter 1996a) in relation to communities of practice (Eckert & McConnell-Ginet 1992) within their historical context. NOTES

1. In Ireland, the term for the language is Irish (or Modern Irish). Gaelic has pejorative connotations (cf. 6 Murchu 1985), suggestive of its earlier history as peripheral to Irish life and as marginal to the powerful institutions that wielded influence over citizens. 2. The concept of internal colonialism, referred to by Nancy Dorian (1981) in her groundbreaking work on the demise of a dialect of Scots Gaelic, concerns the domination of the power structures of a colonizer society over the lives of inhabitants within or near its own boundaries. 3. In fact, the increasing numbers of women working on all levels of media production was remarked on by some of my Irish-language consultants. But women's participation does not come without cost. Irish psychologist Maureen Gaffney has pointed out that women working within the public sphere wrestle with the tensions not only of breaking out of the boundaries of gendered domains particular to Irish society but also of mediating between traditional feminist orientations toward "collectivity-and-inclusivity" and the traditional male "world of hierarchies" (1996:178). She includes women working in business, trade unions, and the media in her analysis. 4. The practice also reflects a more inclusive view toward preservation that has developed in the latter half of the century after a realization that an Irish-only stance was not practical. Angela Bourke writes, "For a hundred years after the famine, Irish people behaved as though they could afford only one language; as though they had to choose between Irish and English: material poverty translated into cultural frugality" (1997:66-67). 5. Compulsory Irish in school has succeeded in making nearly every citizen familiar with the language, as Reg Hindley (1990) and others describe in detail, but has not been able to expand domains for its use.

From Folklore to "News at 6"

385

6. Indeed, during one week in June 1997 there were nearly a half-dozen open positions in Irish-language media, from reporter to technician to desktop publishing worker. 7. Irish has no single indigenous standard; its official synthetic standard, An Caighdean, is taught in schools and to nonnative speakers, but even the official standard does not have the prestige of the Gaeltacht varieties. 8. Because women's representation in the media is not entirely on par with men, particularly in technical areas (Majella Nf Chriochain, personal communication), the Women on Air project was launched in 1996. The aim of the project, jointly sponsored by University College Galway, the Independent Radio and Television Commission, and Connemara Community Radio, is to promote equal opportunity for women in independent radio venues. This support is expected to come through policy development, research (a first-ever study on women's underrepresentation in broadcasting was expected to be completed in late 1997), and training, including a foundation course in radio skills and a diploma in applied communications for radio, which graduated sixteen participants in its inaugural class in May 1997. 9. Underlined words are English in the Irish text (due to considerations of length, only English translations are provided here). Brackets indicate unintelligible words or comments on manner of delivery, as well as brief stretches of content summary. Whereas in Englishlanguage discourse data, lines of talk are generally divided according to intonation unit, this is impossible to render in the present data because intonation plays a different role in Irish (Cotter 1996c). Instead, the line breaks in the data are organized roughly according to clauseor sentence-level boundaries when intonation or discourse-level "chunking" (such as selfinterruption) fails to provide a reasonable alternative. Intonational features are indicated by punctuation (a period meaning falling intonation to low level and a short pause; a comma meaning falling intonation to mid level and a short pause; and a question mark meaning rising intonation). Overlaps or simultaneous speaking are indicated by I, and latching (or no gap between speakers) is indicated by / or V Lengthened phonemes are followed by :. REFERENCES

Bell, Allan (1991). The language of news media. Oxford: Blackwell. Bourke, Angela (1997). Language, stories, healing. In Paul Brennan & Catherine de Saint Phalle (eds.), Arguing at the crossroads: Essays on changing Ireland. Dublin: New Island Books, 58-76. Brody, Jill (1995). Orality, radio, and literacy in the intertextual gap. Paper presented at the annual meeting of the Linguistic Society of America, New Orleans. Bucholtz, Mary (1996). Black feminist thought and African American women's linguistic practice. In Victoria L. Bergvall, Janet M. Bing, & Alice F. Freed (eds.), Rethinking language and gender research: Theory and practice. London: Longman, 267-290. Castillo, JosefinaM. (1994). Waves of change: The experience of Campesinas Unidas de Veracruz. In Mary Bucholtz, A. C. Liang, Laurel A. Sutton, & Caitlin Hines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 79-85. Coates, Jennifer (1989). Gossip revisited: Language in all-female groups. In Jennifer Coates & Deborah Cameron (eds.), Women in their speech communities. London: Longman, 94-122. Cotter, Colleen (1996a). Irish on the air: Media, discourse, and minority-language development. Ph.D. diss., University of California, Berkeley. (1996b). Engaging the reader: The changing use of connectives in newspaper discourse. In Jennifer Arnold et al. (eds.), Sociolinguistic variation: Data, theory, and analysis. Stanford: Center for the Study of Language and Information, 263-278.

386

IDENTITY AS IMPROVISATION

(J996c). Systems in contact: Focus in Irish and English. In Alison Henry, Martin Ball, & Margaret McAliskey (eds.), Proceedings of the International Conference on Language in Ireland. Belfast Working Papers in Language and Linguistics 13. Belfast: University of Ulster, 85-116. (in press). Language and media: Five facts about the Fourth Estate. In Rebecca Wheeler (ed.), The workings of language: From prescription to perspective. Westport, CT: Praeger. Craig, Colette G. (1992). Miss Nora, rescuer of the Rama language: A story of power and empowerment. In Kira Hall, Mary Bucholtz, & Birch Moonwomon (eds.), Locative power: Proceedings of the Second Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 80-88. Dorgan, Howard (1993). The airwaves of Zion: Radio and religion in Appalachia. Knoxville: University of Tennessee Press. Dorian, Nancy (1981). Language death: The life cycle of a Scottish Gaelic dialect. Philadelphia: University of Pennsylvania Press. (ed.) (1989). Investigating obsolescence: Studies in language contraction and death. Cambridge: Cambridge University Press. (1991). Surviving the broadcast media in small language communities. Educational Media International 28:134-137. Eckert, Penelope, & Sally McConnell-Ginet (1992). Think practically and look locally: Language and gender as community-based practice. Annual Review of Anthropology 21:461-490. Foley, Timothy P. (1997). Public sphere and domestic circle: Gender and political economy in nineteenth century Ireland. In Margaret Kelleher & James H. Murphy (eds.), Gender perspectives in nineteenth century Ireland: Public and private spheres. Dublin: Irish Academic Press, 21-35. Freed, Alice (1996). Language and gender research in an experimental setting. In Victoria L. Bergvall, Janet M. Bing, & Alice F. Freed (eds.), Rethinking language and gender research: Theory and practice. London: Longman, 54-76. Gaffney, Maureen (1996). The way we live now. Dublin: Gill/Macmillan. Goodwin, Marjorie Harness (1993). Accomplishing social organization in girls' play: Patterns of competition and cooperation in an African American working-class girls' group. In Susan Tower Hollis, Linda Pershing, & M. Jane Young (eds.), Feminist theory and the study of folklore. Urbana: University of Illinois Press, 149-165. Greenwood, Alice (1996). Floor management and power strategies in adolescent conversation. In Victoria L. Bergvall, Janet M. Bing, & Alice F. Freed (eds.), Rethinking language and gender research: Theory and practice. London: Longman, 11-91. Hindley, Reg (1990). The death of the Irish language: A qualified obituary. London: Routledge. Hinton, Leanne (1994a). The role of women in Native American language revival. In Mary Bucholtz, A. C. Liang, Laurel A. Sutton, & Caitlin Hines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 304—312. (1994-b). Flutes oj fire: Essays on California Indian languages. Berkeley: Heyday Books. Hollis, Susan Tower, Linda Pershing, & M. Jane Young (eds.) (1993). Feminist theory and the study of folklore. Urbana: University of Illinois Press. Jaffe, Alexandra (1994). Media, language and identity on Corsica. Paper presented at the annual meeting of the American Anthropological Association, Atlanta. Kelleher, Margaret, & James H. Murphy (eds.) (1997). Gender perspectives in nineteenth century Ireland: Public and private spheres. Dublin: Irish Academic Press. Kiberd, Declan (1996). Inventing Ireland. Cambridge, MA: Harvard University Press. Krauss, Michael (1992). The world's languages in crisis. Language 68(l):l-42.

From Folklore to "News at 6 "

387

Labov, William (1990). The intersection of sex and social class in the course of linguistic change. Language Variation and Change 2:205—254. Leitner, Gerhard (1980). BBC English and Deutsche Rundfunksprache: A comparative and historical analysis of language on the radio. International Journal of the Sociology of Language 26:75-100. Mills, Kay (1988). A place in the news. New York: Dodd, Mead. Milroy, James, & Lesley Milroy (1991). Authority in language: Investigating language standardisation and prescription. London: Routledge. 6 Donaill, Eamonn (1995). Amharc Neamleanta ar staid na Gaeilge faoi lathair (An unscholarly look at the current state of the Irish language). In Seosamh O Murchu, Micheal O Ceariiil, & Antain Mag Shamhrain (eds.), Oghma, Vol. 7. Dublin: Foilseachain Oghma, 57-65. 6 Murchu, Mairtfn (1985). The Irish language. Dublin: Department of Foreign Affairs/Bord na Gaeilge. 6 Riagain, Padraig (1992). Language maintenance and language shift as strategies of social reproduction: Irish in the Corca Dhuibhne Gaeltacht: 1926-1986. Baile Atha Cliath: Institiiiid Teangeolaiochta Eireann. Pease, Edward C., & Everette E. Dennis (eds.) (1995). Radio: The forgotten medium. New Brunswick, NJ: Transaction. Schiffrin, Deborah (1987). Discourse markers. Cambridge: Cambridge University Press. Shuman, Amy (1993). Gender and genre. In Susan Tower Hollis, Linda Pershing, & M. Jane Young (eds.), Feminist theory and the study of folklore. Urbana: University of Illinois Press, 71-88. Smolicz, Jerzy J. (1992). Minority languages as core values of ethnic cultures: A study of maintenance and erosion of Polish, Welsh, and Chinese languages in Australia. In Willem Ease, Jaspaert Koen, & Sjaak Kroon (eds.), Maintenance and loss of minority languages. Amsterdam: John Benjamins, 277-306. Spitulnik, Debra (1994). Code-mixing and ideologies of hybrid vs. "pure" language use. Paper presented at the annual meeting of the American Anthropological Association, Atlanta. Thomason, Sarah Grey, & Terrence Kaufman (1988). Language contact, creolization, and genetic linguistics. Berkeley: University of California Press. Troemel-Ploetz, Senta (1992). The construction of conversational equality by women. In Kira Hall, Mary Bucholtz, & Birch Moonwomon (eds.), Locating power: Proceedings of the Second Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 581-589. Watson, Seosamh (1989). Scottish and Irish Gaelic: The giant's bed-fellows. In Nancy C. Dorian (ed.), Investigating obsolescence: Studies in language contraction and death. Cambridge: Cambridge University Press, 41-59. Woolard, Kathryn (1989). Language convergence and language death as social processes. In Nancy C. Dorian (ed.), Investigating obsolescence: Studies in language contraction and death. Cambridge: Cambridge University Press, 355-367.

20

M A R J O R I E HARNESS GOODWIN

Constructing Opposition within Girls' Games

.:•:•••• ::;: ince their beginnings in the 1970s, studies of women's language have IB,,, proliferated stereotypes, positing deficit views of female interaction pat;;: Nl; terns and supporting the notion that women's "essential nature" is |:;,. ,f' nonconfrontational and apolitical. These studies can be faulted on two counts: First, they neglected the diversity of ways of speaking in endogenous settings where women live their lives; second, researchers assumed white middle-class American women's speech to be the norm and found women's speech deficient with respect to men's speech (Foster 1995; Henley 1995; Kramarae 1990; Morgan, chapter 1, this volume). Many early studies viewed women as victims forced to act in weak, passive, irrational, ineffective ways. Looking largely at women in cross-sex interaction rather than in same-sex groups, this model ignored the complex ways women interact with one another, as well as the variety of codes from which they may select. Cultural feminists' alternative approaches celebrated the distinctiveness of women's language and highlighted women's supportive interactional styles (for literature reviews see Bing & Bergvall 1996; Freed 1995; Freeman & McElhinny 1995). Women's cooperative talk with its "underlying esthetic or organizing principle of 'harmony' is emphasized in this model" (Kalcik 1975:6). Women complete each other's turns, repeat or paraphrase each other's contributions, and talk simultaneously but not in competition for the floor, preserving equal status and maintaining social closeness (Coates 1991; Falk 1980). According to Senta Troemel-Ploetz (1992:588), "equality among speakers" is achieved by mitigating orders and "toning down and camouflaging dominant speech acts." Jennifer Coates argues that women's talk in the private sphere favors "linguistic strategies which emphasize solidarity rather than status" (1994:79), whereas men's talk in the public sphere is more "information388

Constructing Opposition within Girls' Games

389

focused and adversarial in style" (1994:78). These models contrast women's cooperative talk with men's competitive conversation. Beyond dualism and master categories Early work viewed gender as a master category to be correlated with speech style without regard for other variables that might index social identity, such as ethnicity, social class, age, or sexuality.1 Major problems with such an approach have been outlined by Penelope Eckert: "Gender does not have a uniform effect on linguistic behavior for the community as a whole, across variables, or for that matter for any individual. Gender, like ethnicity and class and indeed age, is a social construction and may enter into any of a variety of interactions with other social phenomena".2 Studies of women's norms of speaking by linguistic anthropologists in Malagasy (Keenan 1974) and Gapun, New Guinea (Kulick 1992,1993) have countered the idea that women are universally "more polite." Recent studies of the speech of women of color in the United States have shown the value of examining social class and age in conjunction with ethnicity; for example, research on the language of Latina women (Galindo 1992, 1994; Galindo & Gonzales Velasquez 1992; Mendoza-Denton 1996, chapter 14, this volume) has challenged stereotypic formulations of Latina women's speech as powerless.3 As studies began to "look more locally" at how speech is used within "communities of practice" (Eckert & McConnell-Ginet 1992; Lave & Wenger 1991; McElhinny 1995), a more complex view emerged. Recently, research conducted in African American speech communities has called into question the earlier focus on unemployed males' street talk as the authentic African American English speech style. And sociolinguistic studies have begun to document the rich codeswitching that takes place among African Americans (Foster 1995; Morgan 1996) and among Spanish speakers (Galindo & Gonzales Velasquez 1992; Galindo 1994; Gonzales Velasquez 1992; Urciuoli 1991; Woolard 1997; Zentella 1997) when selecting from alternative language varieties. The relationship between language, ethnicity, and identity has been problematized as researchers seek a more contextualized understanding of how people mobilize linguistic resources for their interactive projects. Researchers who look at the relationship of ethnicity, nationalism, and language (Gumperz 1982; Jackson 1974; Le Page & Tabouret-Keller 1985) no longer assume simple one-to-one correlations between language and speech community. Rather, through language choice in multilingual situations people actively "project their view of themselves in relation to the universe in which they feel they live and the social structures it contains" (Le Page & Tabouret-Keller 1985:247). The selection of a language variety constitutes an act of identity—"a very overt symbolization of ourselves and of our universe" (1985:248). As John Gumperz and Jenny Cook-Gumperz (1982) argue, in addition to marking identity through language choice, speakers convey ethnicity and group membership through communicative conventions: different cultural assumptions about the presentation of self, ways of structuring information, and types of prosody and contextualization cues (1982).

390

IDENTITY AS IMPROVISATION

This dynamic view of identity also underlies recent studies of "the anthropology of performance" (Turner 1985, 1988), which draws on Erving Goffman's dramaturgical perspective on human experience and the processual view of social relations elaborated by Sally Moore (1978).4 Moore (1978:39) argues that social life "proceeds in a context of an ever-shifting set of persons, changing moments in time, altering situations and partially improvised interactions," in which "established rules, customs, and symbolic frameworks exist, but they operate in the presence of areas of indeterminacy, or ambiguities, inconsistencies, and often contradictions." Such a perspective informs recent gender studies as well. The notion of gender-bending or gender variance (Jacobs & Cromwell 1992) has emerged as important in studies of gender as cultural performance (Butler 1990). Examples include Sherry Turkle's (1995) vignettes of gender blurring and multiple identities on the Internet; Anna Livia's study (chapter 17, this volume) of nongendered pronoun systems in feminist fiction, and Barrett's investigation (chapter 16, this volume) of how different aspects of identity can become one's "operating culture" (Goodenough 1981) at specific moments within interaction. Repertories of identities are taken as the norm, and people play with the variety of linguistic devices through which identities are indexed. Such ideas are compatible both with Fredric Jameson's (1984) notion that, rather than being a monolithic construct, identity is situated in discrete and often self-contradicting locations and with Anthony Giddens' s (1990) idea that "the self" is part of an ongoing dialogue that individuals sustain with themselves in relation to their changing experiences. Identity and language practice Studies of the relationships between language practices and social identity provide important commentaries on and investigations of women's position in society. As Rusty Barrett (chapter 16, this volume) argues, whereas examples of the speech of African American drag queens might "suggest a form of resistance toward racism and homophobia, they do little to call into question the sexism in American society." And Colleen Cotter's study (chapter 19, this volume) of media discourse in Irishlanguage radio shows that although media discourse promotes cultural identity, it does little to promote women's discourse forms, as Irish identity is filtered through "genderless" rather than female ways of speaking over broadcast media. The ways in which women's talk is manipulated for commercial gain is examined in Mary Bucholtz's study (chapter 18, this volume) of the coconstruction of a fictional community on the shopping channel. Bucholtz's account of how shopping-channel participants build a shared social world is compatible with recent work in social constructionism (Coupland & Nussbaum 1993), a psychological perspective that views all social categories and identities as constituted through discourse (Coupland, Nussbaum, & Grossman 1993). Such a perspective within linguistics builds from work in conversation analysis, whose project is to describe "the procedures by which conversationalists produce their own behavior and understand and deal with the behavior of others" (Heritage & Atkinson 1984:1). Rather than assuming that identity can be easily correlated with social vari-

Constructing Opposition within Girls' Games

391

ables such as ethnicity, social class, gender, or age, conversation analysts argue that identity is shaped moment to moment through the details of how participants interact within activities. Thus Lisa Capps (chapter 4, this volume) shows how the identity of the "irrational woman" develops within storytelling interaction. In pursuing a response from a family member who displays minimal or no responsiveness to her story, the agoraphobic woman Capps studied escalates her story of danger or helplessness, thereby amplifying her deviant status. By using the methodology of conversation analysis, Capps shows how language not only reflects social order but also constitutes it— and simultaneously constitutes a particular psychological condition. The notion of identity developed by Capps and many other authors in this book (and in particular those included within part IV) is one that is fundamentally activityfocused, created through language choices from linguistic repertoires. As Goffman argues, it is coordinated task activity rather than conversation that organizes a great deal of conversational interaction: "A presumed common interest in effectively pursuing the activity at hand, in accordance with some sort of overall plan for doing so, is the contextual matrix which renders many utterances, especially brief ones, meaningful" (1981:143). As participants move through activities, they take on appropriate roles, which constitute occasion-specific identities. Such roles are located within what Goffman has called a "situated activity system": "a somewhat closed, self-compensating, self-terminating circuit of interdependent actions" (1961:96). Occasion-specific identities are those interactional practices proposed and ratified in the course of specifiable activities—for example, in the case of telephone conversations, caller and called, or in a dispute, accuser and defendant. In addition, participants articulate their position or stance with respect to a present project through ritual displays, providing "evidence of the actor's alignment in the situation" (Goffman 1979:1) or footing (Goffman 1981)—one's stance, posture, or projected self. Indeed, for this reason Deborah Tannen (chapter 11, this volume) rejects the term identity altogether in favor of Goffman's concept of display, which suggests that the self is not static but created dynamically in action.5 Participants try on and play with a host of diverse social personae as they navigate through different positions within activities. Dualism in research on children's gendered identities In attempting to provide global differentiations between gender groups, social scientists have often dichotomized differences between girls' and boys' interactions. Sociologists (Lever 1978) and psychologists (Borman & Frankel 1984; Borman & Kurdek 1987; Piaget [1932] 1965; Sutton-Smith 1979) have claimed that girls generally are less concerned than boys with making and arguing about rules. Sociolinguistic and psychological research on female interaction has been deeply influenced by psychologist Carol Gilligan' s theory of moral development, which is based on reports told to a researcher rather than on recorded interaction and ethnography: "This [female] conception of morality as concerned with the activity of care centers moral development around the understanding of responsibility and relationships, just

392

IDENTITY AS IMPROVISATION

as the conception of morality as fairness ties moral development to the understanding of rights and rules" (Gilligan 1982:29; original emphasis). Gilligan argues that girls are primarily concerned with maintaining relationships within intimate social groups. Her dualistic "different voice" argument has been echoed in the work of a number of psychologists (e.g., Leaper 1991; Miller, Danaher, & Forbes 1986). Similarly, according to Kathryn Borman and Lawrence Kurdek (1987:248-250), girls' games stress interpersonal understanding, whereas boys' activity is "structured, rulegoverned, competitive, and team oriented." Both they and Janet Lever (1978) cite jump rope and hopscotch as two quintessential cooperative girls' games which, being characterized by turn-taking, lack the elaboration of rules or differentiated social roles. Such studies conclude that "the play activities of boys are more complex than those of girls, resulting in sex differences in the development of social skills potentially useful in childhood and later life" (Lever 1978:472).6 Studies of the actual playing of girls' games in context defy this essentialist, deficit view of girls' interactive skills. Analyzing girls' play during hopscotch, I found that every move within the game was fiercely scrutinized for its possible categorization as a foul or violation. Girls seemed to delight as much in arguing, reconstructing, and violating the rules as in actually winning. By examining the specific linguistic resources and practices used to build turns at talk we can see how girls actively construct themselves as agents who are responsible for monitoring the social order. These girls defy the standard, essentialized stereotypes of Latinas as the hapless victims of a patriarchal culture (Anzaldua 1987, 1990; J. Moore 1991; Orenstein 1994; Sadker & Sadker 1994), for they are neither passive nor muted. 7 Like the Chicanas studied by Letticia Galindo (1994) and Norma Mendoza-Denton (chapter 14, this volume), the girls in the present study use language creatively and without inhibition. Fieldwork The present data are based on a study of the interaction of bilingual Spanish- and English-speaking girls, primarily second-generation Mexican and Central Americans, in an elementary school located in the Pico Union/Koreatown district near downtown Los Angeles. The school population of 828 students was predominantly Latino (92.2%). Ninety-five percent of the children were eligible for the federal meal program under the Chapter 1 program. Children were transitioned into English reading toward the end of the third grade; fourth and fifth grades were taught primarily in English. The girls in this study were in the second, third, and fifth grades and generally spoke Spanish to one another, except when referring to hopscotch moves or numbers on the squares of the play grid.8 English was thus reserved for a domain-dependent lexicon; terms such as Out! or You 're out and Sorry, all of which meant that a player would forfeit a move, were spoken in English. In addition, when performing a part of the game called ABC (discussed later), girls used English phonology to repeat the letters. As a middle-aged Anglo professor and community outsider, I was obviously different from the children I was observing, but the fact that Lori Cronyn, a colleague

Constructing Opposition within Girls' Games

393

at UCLA, was a teacher at this school and was well known by the children provided an extraordinary opportunity for fieldwork. My previous research on working-class African American children's interactions (Goodwin 1990) provided me with a basis for comparing speech styles; I began to see cross-cultural similarities in children's formulations of disagreement in opposition moves. Additionally, I noticed that, as in my earlier work, the identities most salient to children's interactions during hopscotch were not macrolevel categories such as gender or ethnicity but instead were locally managed identities that emerged from ongoing practices during play.

Occasion-specific identities within hopscotch Play took place after school, during lunch break, and at two recess periods; teachers often extended recess, reasoning that outside the safety of the school grounds, children had little opportunity to play with peers. Before and during school, children often played under the supervision of a coach whose primary duties were to organize children's sports. After school, children whose parents permitted them to remain on the playground could get equipment from a playground supervisor to play billiards (played by both girls and boys, though in same-sex groups), hopscotch (primarily a girls' game in this community), or various ball games (such as dodgeball and handball) on the expansive playground, which covered two city blocks.9 Freed from adult supervision, they could organize their own activities, a situation that provided me an excellent opportunity to observe peer interaction. Girls and boys were in proximity to each other during after-school hours, although they tended to play in same-sex groups. During hopscotch, girls protected the boundaries of their play space from boys' intrusions, delimiting their territory through what Barrie Thorne (1993), following Fredrik Barth (1969), has called "borderwork." When Spanish-speaking boys intruded into girls' space, girls would push them away. With commands such as Get out of the way! girls sanctioned the boys' behavior and marked them as different by switching to English when addressing them. Such spatial considerations are crucial in the investigation of games like hopscotch, which are locatable in a particular area of the playground. Hopscotch is a "situated activity system" (Goffman 1961) involving a physical environment (a grid with nine spaces painted on a cement schoolyard; see figure 20.1) and procedures for moving within it. The object of the game is to be the first player to advance her beanbag from the lowest to the highest square and back again by means of a series of beanbag tosses and player jumps. From behind the start line, each player in turn tosses a beanbag into a square and, without changing feet, jumps to the end of the grid and back again on one foot, avoiding squares where beanbags have been tossed. If the player falls down or steps on a line or outside the appropriate square, she must forfeit her turn. When a player performs an infraction, "Out!" is called. Once a game has begun, girls may either welcome new players, even permitting them to advance their beanbag to the same square theirs is on, or alternatively prevent newcomers from joining. Excluded girls may either protest their denial of entry into the game and wait on the sidelines as peripheral participants commenting on the

394

IDENTITY AS IMPROVISATION

Figure 20.1. The hopscotch grid

play or they may form their own new game on an adjacent grid. Age and friendship are important factors influencing such decision making and alliance formation. In the data analyzed here, a group of fifth-grade girls who initiated the game prevented younger girls' participation as jumpers and ignored their judgment calls about plays. When the older girls went home, the younger girls initiated a new game and became outspoken critics of moves in progress. Although it is commonplace to view a game with respect to its players, girls who observe and assess the game activity are equally relevant to the play. Onlookers are not random spectators; they have a vested interest in finding violations, for they get a chance to jump when ajumper forfeits her turn. As jumpers continually try to have their fouls overlooked, the job of girls on the sidelines is to vigilantly monitor or referee the ongoing action. However, if they are not ratified as valid participants, their judgments will be ignored. A range of alignments to the action in progress is possible. Girls on the sidelines clap, jump up and down, and laugh loudly when other players miss. They seem to get as much pleasure from this activity as from jumping through the grid. Girls playfully tease those preparing to jump (or even in the midst of their jumps), lightly pushing them, spanking them, and trying to unnerve them—for example, pointing to their feet and shouting, "Un raton! Un raton!" (A mouse! A mouse!). In hopscotch, although the game is played to win, it is richly overlaid with multiple framings and textured nuances such as joking, tricking, and taunting. Girls compete for being first in a round of hopscotch. During the course of the game girls will openly brag about their successful playing, sing-chanting, "Voy ganando, voy ganando EY:::::::::::!" (I'm winning! I'm winning!) or "Que bueno, yo voy en el ultimo!" (How terrific, I'm going to the last square!). Although such statements are one way to claim one's relative position (which may be countered in subsequent interaction), trickery is another possibility. From the player's perspective, the most interesting part of the game is seeing what one can get away with, playfully extending possible definitions of the rules to make them fit the circumstances

Constructing Opposition within Girls' Games

395

of the moment. Indeed, for the 20-minute session of hopscotch analyzed here, different girls selected from these alternative strategies for asserting their respective positions. One girl, a newcomer who lacked proficiency in executing some of the moves, proved quite skilled in tricking others and providing her own metacommentary on participation. Players are continuously engaged in testing the limits and boundaries of rules in this way. Participant structure and the sequential structure of out-call sequences In monitoring a player's activity, referees can call a foul in a number of different ways. Perhaps the most common is to yell, "Out!" and provide (1) an account for the out call (Pisaste la ray a! 'You stepped on the line'), accompanied by (2) a demonstration of how it occurred, as in Example (1): (1)

3:57:41 1 Marta: 2 Roxana: 3 Carla: 4 Roxana: 5 Marta: 6 Gloria: 7 Carla: 8 Gloria: 9 Marta:

((jumps with one foot outside grid)) [Out. Lout! °Out. AY:::! ((throws up hands smiling, turning head)) ["HAH HAH! LPisaste la ray a! ((stepping multiple times on line where violation occurred)) You stepped on the line. ((claps hands three times excitedly while laughing)) ((throws beanhag to appropriate next square, ending her turn, and moves back to front of grid))

Commentary on the game provides a way of displaying one's alignment toward the action in progress. During the activity of reacting to a violation, the display of affect occurs in a specific sequential position and is made relevant by practices for performing the out call. The preface is critical in constructing an opposition move, for it states quite literally a stance or footing with regard to the current action. In this example Roxana and Carla's out call and Carla's account (line 7)—You stepped on the line!—is followed by an animated demonstration. Carla intensifies her verbal sanctioning action by stepping several times on the line where the out occurred. Thus in multiple ways she performs the activity of evaluation. In order for her action to count as an appropriate move in the language game of calling out, Carla's actions must be ratified by others present. The sequential organization of moves is important in that players build the game world by assessing whether their actions count as valid moves. In example (1) the referees' call is ratified not only by Gloria, another onlooker, through her laughter (line 6) and clapping (line 8) but also by the jumper Marta. Marta smiles while producing a response cry AY::: (line 5), closing her eyes and turning her head in a posture that displays her own

396

IDENTITY AS IMPROVISATION

humorous "take" on her foul. Through such moves she acknowledges her error and aligns herself with the positions of both onlookers; subsequently (line 9) she moves her beanbag to the next square, effectively marking that her turn has ended. The player's uptake—her response cry marking agreement with the call—contrasts with other possible displays, such as disgust or defiance (exhibited by stomping back to the start line, throwing another's beanbag across the playground, screaming, and so forth), and provides closure. The flagging of a problem leads quickly to an acknowledgment of error; the sequence thus has the character of a remedial sequence (Goffman 1971). The occasion-specific identities of player and evaluator are constructed through these actions. If the jumper rejects an onlooker's judgment, however, the sequence will play out differently. Players frequently feign innocence and ignore out calls, especially when others who will support the player's position are present. In one instance, moments after two onlookers had called someone out, the player resumed her turn. Subsequently, when a first referee asked her coreferee "^Verdad que si esta out, verdad?" (It's true that she's out, right?) the coreferee reversed her own prior ruling: "No se. No me preguntas a mi" (I don't know. Don't ask me). The coreferee thereby aligned with the player's position against the other referee's call, and the player continued her turn. The identity of friend-friend is constructed when, as here, an onlooker sides with the player against a coevaluator's negative assessment.

The shape of out calls: Pitch contours In children's arguments, turn shapes may display "aggravated correction" (M. Goodwin 1983). Whereas in adult forms of disagreement nonsalient intonation is preferred over expressions of disagreement (Yaeger-Dror 1986), children clearly signal opposition. In the next example an argument develops between Carla and Gloria over whose turn it is. When Carla declares, "Ya voy" (I'm going now), usurping Gloria's turn, Gloria states in response, "N'ai:::! Ya voy YO!" (No. I'm going now). (2) 4:11:08

Carla:

Ya voy.

Gloria: N'ai:::! Ya voy YO!

The oppositional turn vividly displays a strong emotional stance or footing, what we might gloss as outraged indignation at the first speaker's despicable behavior. How is this stance made visible? The speaker begins her oppositional turn with a preface, N'ai, announcing an objection to the prior move at the earliest possible opportunity. Moreover, this preface is spoken with a dramatic pitch excursion (see figure 20.2). Although the first speaker's talk is produced at a pitch between 300 and 400 Hz, the oppositional turn N'ai;;;/ leaps quickly and dramatically to 600 Hz. Within the single syllable of the preface, the second speaker's voice leaps from 400 to 600 Hz. The display of outrage, with its associated emotional components, is made visible as an embodied performance through the second speaker's voice and intonation. In addition, although in Spanish subject pronouns are not required, in Gloria's counter the

Constructing Opposition within Girls' Games

397

Figure 20.2. The pitch contour of oppositional turn prefaces. SOURCE; Goodwin 1998

pronoun YQ/ is appended to the end of the utterance and emphasized through "contrastive stress" (Ladd 1980:78); YQ/ is produced at 500 Hz. Dwight Bolinger (1983) has hypothesized a relationship between higher vocal pitch and displays of increased excitement. In the present example, affective intensity is marked through vowel lengthening, singsong intonation, and raised volume. (The pitch at the beginning of example (2) is above 600 Hz; the girls' normal voice range is between 250 and 350 Hz.) However, pitch height does not function as an isolated, decontextualized display. Instead it becomes visible as a specific, meaningful event, by virtue of its embedding within a particular sequence of action. Like the turn preface, the squeal of outrage is also indexically tied to the immediately prior action, which constitutes the point of departure for the display of opposition. The second speaker builds her moves within a field of meaning that has been brought into existence by the conditional relevance (Schegloff 1968) of the prior action. On the level of sound structure, pitch height becomes a salient action because it vividly contrasts with the preceding talk. In addition to pitch, participants may utilize posture and gesture to accuse another girl of having landed on a line while making a jump in hopscotch (M. Goodwin 1998). For example, out calls are frequently accompanied by pointing of the index finger, as shown in figure 20.3. Turn construction in opposition moves Sequential slots for the production of relevant responses provide participants with places where they can use a range of embodied activity to build appropriate action.

398

IDENTITY AS IMPROVISATION IP? HiiS,

Carla OUT! OUT! Figure 20.3. Extended hand point in out call. SOURCE: Goodwin 1998

Polarity expressions such as NO (example 3), response cries (examples 4 and 5)— terms such as AY:!, EY!, Ou:!, or Ah:!—and negative person descriptors (M. Goodwin 1990) such as chiriona (cheater) constitute important components of the turn, providing other ways to indicate the referee's alignment or affective stance toward a prior action: (3)

Gloria: ((jumps from square 2 to Problematic Move square I, changing feet)) Polarity Expression + Negative Person Descriptor Carla: NO CHIRIONA! No, cheater! YA NO SE VALE AST. Explanation That way doesn't count anymore! (4)

Gloria: ((takes a turn out of turn)) Problematic Move Response Cry + Negative Person Descriptor Carla: AY: TU CHIRIONA! Hey, you cheater! Explanation EH NO PISES AQUI Hey, don't step here. PORQUE AQUI YO VOY! Because I'm going here.

Constructing Opposition within Girls' Games

399

(5)

Gloria: ((Jumps from square 3 to 2, Problematic Move changing feel)) Carla: EY::!! CHIRIONA! Response Cry + Negative Person Descriptor MIRA! Hey! Cheater! Look! TE VEN1STES DE AQUI AST! Explanation You came from here like this, ((demonstrating how Gloria jumped, changing feet))

With negative person descriptors, referees argue not simply that an infraction has occurred but that the player's action is morally wrong. The term chiriona, derived from the English word cheat plus -ona, a Spanish agentive nominalizer or intensifier, provides an explicit characterization of the person whose move is opposed.10 By using such a term, a judge argues not simply that an infraction has occurred but that the person who committed the foul is accountable for its occurrence. Following the opposition preface, a referee provides a reason why the move is invalid, often through a demonstration. Unlike the delayed disagreement in adult conversation (Pomerantz 1984; Sacks [1973] 1987), here the girls, through intonation and gestures (such as extended hand points), display opposition in no uncertain terms. The direct initiation of remedial sequences is consistent with other researchers' findings regarding children's practices of other-correction (Field 1994; Maynard 1986).

Breaking the rules Some moves are deliberate violations; indeed, breaking the rules is very much a part of the game. This fact is illustrated in example (6), which occurs as Sandra, who joined the others after some 6 minutes of play, learns the rules of ABC. Looking toward the other players as she takes her third step a bit larger than permitted, the jumper herself keys her move as inappropriate through her laughter. The referees counter her move with a marker of polarity NO:: (line 2), a response cry AY::: (line 3), and opposition turns containing negative person descriptors: NO CHIRIONA! (line 4) and Cheater! (line 7). Sandra persists in taking larger steps than are allowed. A referee shows her how she must place her feet precisely on the line, with each step aligning heel and toe of opposite feet, in contrast to the large steps she is taking. Her moves are countered again with NO::: (line 9) and AY:::! (line 11), this time accompanied by a justification (lines 13-16): QUE T1ENES QUE METERTE EN LA RAYA DE AQUI LOS DOS JUNTITOS. AL OTRO PIE NINA! (You have to put yourself on this line and put each foot close behind the other one, girl!). The verbal statement is accompanied by enactments of precisely how to place one's feet on the grid. (6)

4:05:22 ((Sandra has just been instructed In how to take baby steps according to ABC rules, putting the heel of one shoe against the toe of the other. Currently she is deliberately taking larger steps than permitted and laughing))

400

IDENTITY AS IMPROVISATION

1 2 3 4

Sandra: A(hh) B, C (h) ((smiling)) Gloria: NO [(lowers lop part of body)) Carla: [AY::: ((playfully giving Sandra a spank)) Carla: NOCHIRIOtNA! No, cheater! 5 Sandra: [Okay. 6 Sandra: [A 7 Gloria: [Cheater! 8 Sandra: B, C.= ((taking big steps)) 9 Gloria: NO::: ((body lowers dramatically)) 10 Sandra: ((smiles widely)) 11 Carla: AY:::! 12 Sandra: [Hih hih! 13 Carla: LQUE TIENES QUE METERTE You have to put yourself 14 EN LA RAYA on this line

IDE AQUI

15 16 17

Sandra: bkay! Carla: LOS DOS JUNTITOS. and put each foot close AL OTRO PIE NINA! behind the other one, girl!

The moves display various forms of opposition through their markers of polarity (No in lines 2, 4, 9), response cries (Ay!, lines 3, 11), negative person descriptors (CHIRIONA, line 4; Cheater!, line 7), and increased volume. At the same time the sequence is keyed with laughter by the jumper, Sandra. As Sandra takes a deliberately expansive step, Gloria lowers the top part of her body and produces a loud NO (line 2) and Carla pushes her out of the position at the start of the grid and shows her the appropriate way to take a turn. When Sandra takes even larger steps, Gloria displays her exasperation by lowering her torso as she states "No" (line 9). The interaction develops a multitextured ethos in which the judges playfully chide Sandra for her performance. The same players who referee others' rule violations may themselves perform violations moments later. In (7), Carla, who 6 minutes earlier had critiqued the way Sandra was taking big rather than small steps, now attempts to rewrite the rules. Proposing a hypothetical version of the rules, she boasts that she is going to take four rather than three baby steps when she does ABC, arguing that the square to which she must advance (nine, the farthest square) is at some distance from her (estd mas lejos). (7)

4:11:00 1 Carla: 2

Ahora voy a hacer (.) cuatro ABC. Now I'm going to take Four ABC steps. Gloria: NO! Al cuatro no! No! Up to four! No!

Constructing Opposition within Girls' Games 3 Carla: 4 Carla:

401

Tiene que ser cuatro porque It's got to be four because ya esta mas lejos. it's farther away.

Extensive negotiation and debate occurs. Far from attempting to avoid conflict, through forms of aggravated correction the girls here actively seek it out. Various keyings occur as coparticipants transform their affective alignment toward the game in different ways throughout its course. A framework for comparison Because many different groups play hopscotch, it provides an invaluable research tool for documenting cross-cultural variation in language use. Remarkable similarities can be observed in the way that Pico Union children and African American working-class children in rural South Carolina structure their opposition moves. As with the Pico Union children, for example, among the South Carolina children Out is yelled immediately after a foul (example 9): (9)

10:47:26 Lucianda: Joy: Crystal: Lucianda:

((puts foot in square with token)) You [but. ((pointing toward jumper)) [.Out! I'm out.

Similarly, in both groups bolder, more authoritative stances occur when the judge not only proclaims the jumper out but intensifies her action through escalation of volume, physical pointing, and an explicit statement that a violation has occurred. Example (10) illustrates this pattern among the South Carolina children: (10) 10:53:30-37 Crystal: ((lands with foot in a square with a token on it)) Lucianda: YOU OUT. YOU TOUT! ((pointing)) Crystal: [Why. Lucianda: YOUR BO- YOUR FEET IN NUMBER FIVE!

Turns that begin with response cries are also similar in both groups; the African American out call in example (11) is prefaced by Ah: rather than the Spanish Ay and is followed by an out call, as well as an account: (11) 11:15:13 Lucianda: Vanessa: Crystal: Vanessa:

((jumps stepping on lines)) Ah: [Lucianda. l_Out! Out- You step between the line. Not in it.

402

IDENTITY AS IMPROVISATION

Such authoritative stances are not unique to these two groups; I have observed similar turn shapes in arguments during hopscotch sessions among children of the fifth-grade ESL class mentioned in note 9. The population was roughly equivalent to the Pico Union school, and included children from Azerbaijan, Saudi Arabia, China, Korea, Vietnam, and Mexico." By contrast, in my fieldwork among middle-class white girls from Columbia, South Carolina, children, rather than making definitive out calls, used hedges (/ think, sort of, and so on) in moves such as / think that's sort of on the line though or You accidentally jumped on that; but that's okay (M. Goodwin 1998). Much more work is required to sort out the effects of ethnicity, age, and social class on norms of speaking.12 Alhough I have presented the materials here with respect to a particular context, that of Spanish-speaking children in Los Angeles, I have not demonstrated the grounds for the relevance of such categorizations in the talk of the children themselves. Many other explanations for behavior are possible, including age, gender, one's newness to the game, and friendship alliances. Similar turn structures for performing oppositional moves can be observed in the structure of interaction within other groups. Rather than focusing exclusively on the peculiarities of particular language-based groups, T prefer to be open to the possible universality of linguistic resources that children of both genders build upon to ratify or disconfirm particular visions of the self and identity (Capps & Ochs 1995).

Conclusion Although anthropologists have documented variations on what are considered the canonical features of female speech, dichotomous views of female and male language persist, echoing Gilligan' s (1982) view that girls speak in a unitary, nonconfrontational form of "different voice," one more concerned with an ethic of care than with the rules of the game. In line with such an essentialist position, the accepted wisdom on girls' play maintains that turn-taking games such as hopscotch are simple grids for the rotation of players through space, limiting girls' experience because they reputedly restrict dispute. As Lever states, Girls' turn-taking games progress in identical order from one situation to the next: prescriptions are minimal, dictating what must be done in order to advance. Given the structure of these games, disputes arc not likely to occur. . . . Because girls play cooperatively more than competitively, they have less experience with rules per se, so we should expect them to have a lesser consciousness of rules than boys. (1978:479)

Lever's assessment of girls' activity echoes pronouncements of Jean Piaget, who argues that "the most superficial observation is sufficient to show that in the main the legal sense is far less developed in little girls than in boys" ([1932] 1965:77). According to Piaget, the game of hopscotch is "very simple and never presents the splendid codification and complicated jurisprudence of the game of marbles." He argues that few girls play marbles, and when they do they are more concerned with "achieving dexterity at the game than with the legal structure."

Constructing Opposition within Girls' Games

403

Such positions about girls' lack of interest in legal structure are unfounded and resemble essentialist accounts of women's language and psychology. The range of experiences encompassed in the game world of hopscotch seriously challenges characterizations of girls' play as limiting their development of social skills (Lever 1978). While playing hopscotch, girls are not simply playing by the rules but playing with the rules, and while judging the performances of their peers they are acquiring rhetorical and negotiating skills important for a range of circumstances in later life. Close ethnographic investigation of girls' disputes during play shows that girls draw from a discourse repertoire (M. Goodwin 1990) and can incorporate cooperation and competition, legalistic language and laughter, within the interaction (Hughes 1988, 1993; Kyratzis 1992, 1994; Sheldon 1992, 1993). Footings or alignments to the action in progress shift as new positions are taken up. Stance is conveyed through body posture and affective intensity (Bradac, Mulac, & Thompson 1995) or highlighting (C. Goodwin 1994), as indicated through pitch leaps, vowel lengthening, and raised volume. With new methods to document language use in social encounters, we no longer need to rely on a priori dualistic categories or interviews that yield information about language rather than linguistic practices. We are positioned to hear multivocality within women's and girls' speech across diverse groups and can analyze identity as shifting rather than static—"an activity or performance" rather than an attribute (McElhinny 1998:469). Arguing against a dualistic view of language and gender, Nancy Henley states that "a repeated lesson is that we must conceive of communication as socially and culturally situated action that is not universally determined by simplistically conceived gender, socialization, role, or personality" (1995:385). Through close analysis of talk in interaction we can show how individuals propose and demonstrate to each other the relevance of particular features of their identity operative on specific occasions of use. APPENDIX: TRANSCRIPTION CONVENTIONS

Data are transcribed according to the system developed by Gail Jefferson and described in Sacks, Schegloff, & Jefferson (1974). Cutoffs:

A dash (-) marks a sudden cutoff of the current sound.

Emphasis:

Underlining and boldface indicate some form of emphasis.

Overlap bracket:

A left bracket ([) marks the point at which the current talk is overlapped by other talk.

Lengthening:

Colons (::) indicate that the sound immediately preceding has been noticeably lengthened.

Intonation:

Punctuation symbols are used to mark intonation changes rather than as grammatical symbols. A period indicates a falling contour, a question mark indicates a rising contour, and a comma indicates a falling-rising contour.

Latching:

Equal signs (=) indicate latching—that is, there is no interval between the end of a prior turn and the start of a next piece of talk.

404

IDENTITY AS IMPROVISATION

Inbreath:

An h preceded by an asterisk (*h) marks an inbreath.

Comments:

Double parentheses (()) enclose material that is not part of the talk being transcribed. They frequently indicate gesture or body position.

Silence:

Numbers in parentheses (0.6) mark silences in seconds and tenths of seconds.

Increased volume:

Capitals (CAPS) indicate increased volume.

NOTES

I would like to acknowledge the following people: Salome Santos and Carla Vale assisted in translating text; Mary Bucholtz, Patrick Gonzales, Roberta Chase-Borgatti, Alicia de Myhrer, Sally Jacoby, Pat Mason, Norma Mendoza-Denton, Marjorie Faulstich Orellana, Manny Schegloff, and Malcah Yaeger-Dror provided many useful comments; Chuck Goodwin helped in all stages of development. This research would not have been possible without Lori Cronyn, who introduced me to the teachers, principal, and children of the school where the study was conducted. I have benefited in countless ways through talks with her about children, schooling, language, and community in Pico Union. 1. On various notions of identity see Amelie Rorty and David Wong (1990). I am not using the term identity to refer to an aspect of one's individual nature but rather to aspects of the self that make a difference in how one conducts oneself. 2. Similarly, Ben Rampton (1995:486), countering what he terms "an absolutist view" of ethnicity, gender, and social class as "a discrete, homogeneous and fairly static cultural essence," argues that we need to look at how class, gender, or ethnicity is "activated in different ways in different contexts." 3. Marjorie Orellana (chapter 3, this volume) discusses ways in which, through "critical pedagogy," working-class Chicano children in a Los Angeles bilingual classroom learn to challenge the constraints of racism and poverty (while, however, less frequently questioning stereotypic gender roles). 4. Goffman (1971:62) defines ritual as "a perfunctory, conventionalized act through which an individual portrays his respect and regard for some object of ultimate value to that object of ultimate value or to its stand-in." He notes (1971:63) that although in contemporary society we rarely practice rituals to stand-ins for supernatural entities, we do display ritualistic behavior toward one another, "attesting to civility and good will on the performer's part and to the recipient's possession of a small patrimony of sacredness." By way of contrast, Turner (1988:75) defines ritual as "a transformative performance revealing major classifications, categories, and contradictions of cultural processes." He states that although he is concerned with "the performance of a complex sequence of symbolic acts," Goffman considers ritual "a standardized unit act, which may be secular as well as sacred." 5. The notion of stance is also developed in work on language and gender by Elinor Ochs (1992) and by Tannen (chapter 11, this volume). 6. Lever's findings are recycled in most literature on children and gender. See, for example, Evelyn Pitcher and Lynn Schultz (1983). 7. To take one example of this essentialized perspective, Peggy Orenstein, quoting the 1992 AAUW report on girls and education, states: "Latina girls report the greatest plunge in self-esteem of any girls surveyed. . . . In their teenage years, they have a more negative body image, are at greater risk of attempting suicide, and report higher levels of emotional stress— anxiety, depression, nervousness, insecurity, or exhaustion—than any other group of children, male or female, of any race or ethnicity" (1994:199).

Constructing Opposition within Girls' Games

405

8. In a May 1996 study of hopscotch at a comparable school in Columbia, South Carolina, with Spanish speakers from Mexico and Puerto Rico who had been in the United States for only a year, I found that children played the game almost exclusively in Spanish. 9. In other settings both girls and boys play hopscotch. In the Columbia study, I observed after-school and recess play in a magnet school. African American, Korean, Japanese, Saudi Arabian, and Mexican boys knew how to play the game. Moreover, girls and boys in a small ESL class of six students played hopscotch together. 10. Shana Poplack (1980) argues that codeswitching does not generally occur at a morpheme boundary unless one of the morphemes has been integrated phonologically into the language of the other. Norma Mendoza-Denton has pointed out to me that this example shows how the bilingual phonology of the children operates: The speaker begins with the English word cheater and codeswitches at a morphological boundary in the middle of the word by changing the /t/ of cheat to /r/. Although the vowel quality is primarily Spanish, the word has an English phonological process operating within it, with the intervocalic flapping of /t/. 11. Eighty-nine percent of the children in the magnet school received free lunches, and 10 percent were eligible for lunch at a reduced fee. 12. Working-class white children in the Baltimore community studied by Peggy Miller (1986) are socialized to be assertive when required to defend themselves. Donna Eder (1990:82) similarly argues that for the working- and lower-class white girls she studied "'toughness' is more highly valued and there is less concern about 'politeness.' " In contrast, the principal of the Columbia school where middle-class children's mitigated responses were observed actively promoted an ideology of conflict avoidance; such an ideology was consistent with the norms of the Unitarian Universalist Church, which two of the four girls in the study attended. REFERENCES

Anzaldiia, Gloria (1987). Borderlands / La frontera: The new mestiza. San Francisco: Spinsters/ Aunt Lute. (ed.) (1990). Making face, making soul / Hacienda caras. San Francisco: Aunt Lute. Barth, Fredrik (1969). Introduction. In Fredrik Barth (ed.), Ethnic groups and boundaries. Boston: Little, Brown, 9-38. Bing, Janet M., & Victoria L. Bergvall (1996). The question of questions: Beyond binary thinking. In Victoria L. Bergvall, Janet M. Bing, & Alice F. Freed (eds.), Rethinking language and gender research: Theory and practice. London: Longman, 1—30. Bolinger, Dwight (1983). Intonation and gesture. American Speech 58:156—174. Borman, Kathryn M., & Judith Frankel (1984). Gender inequities in childhood: Social life and adult work life. In Kathryn M. Borman, Daisy Quarm, & Sarah Gideonse (eds.), Women in the workplace: Effects on families. Norwood, NJ: Ablex, 113-135. Borman, Kathryn M., & Lawrence A. Kurdek (1987). Grade and gender differences in and the stability and correlates of the structural complexity of children's playground games. International Journal of Behavioral Development 10(2):241— 251. Bradac, James J., Anthony Mulac, & Sandra A. Thompson (1995). Men's and women's use of intensifiers and hedges in problem-solving interaction: Molar and molecular analyses. Research on Language in Social Interaction 28(2):93—116. Butler, Judith (1990). Gender trouble: Feminism and the subversion of identity. New York: Routledge. Capps, Lisa, & Elinor Ochs (1995). Constructing panic: The discourse of agoraphobia. Cambridge, MA: Harvard University Press.

406

IDENTITY AS IMPROVISATION

Coatcs, Jennifer (1991). Women's cooperative talk: A new kind of conversational duet? In Claus Uhlig & Rtidiger Zimmermann (eds.), Proceedings of the Anglistentag 1990 Marburg. Tubingen: Max Niemeyer Verlag, 296-311. (1994). The language of the professions: Discourse and career. In Julia Evetts (ed.), Women and career: Themes and issues in advanced industrial societies. London: Longman, 72-86. Coupland, Nikolas, & Jon F. Nussbaum (eds.) (1993). Discourse and life span identity. Newbury Park, CA: Sage. Coupland, Nikolas, Jon F. Nussbaum, & Alan Grossman (1993). Introduction: Discourse, selfhood and the lifespan. In Nikolas Coupland & Jon F. Nussbaum (eds.), Discourse and life span identity. Newbury Park, CA: Sage, x-xxxviii. Eckert, Penelope (1989). The whole woman: Sex and gender differences in variation. Language Variation and Change 1:245-267. Eckert, Penelope, & Sally McConnell-Ginet (1992). Think practically and look locally: Language and gender as community-based practice. Annual Review of Anthropology 21:461-490. Eder, Donna (1990). Serious and playful disputes: Variation in conflict talk among female adolescents. In Allen D. Grimshaw (ed.), Conflict talk: Sociolinguistic investigations oj arguments in conversations. Cambridge: Cambridge University Press, 67-84. Falk, Julia (1980). The conversational duet. In Bruce R. Caron et al. (eds.), Proceedings of the Sixth Annual Meeting of the Berkeley Linguistics Society. Berkeley, CA: Berkeley Linguistics Society, 507-514. Field, Margaret (1994). On the internalization of language and its use: Some functional motivations for other-correction in children's discourse. Pragmatics 4(2):203-220. Foster, Michele (1995). "Are you with me?": Power and solidarity in the discourse of African American women. In Kira Hall & Mary Bucholt/, (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 329-350. Freed, Alice (1995). Language and gender. Annual Review of Applied Linguistics 15:3—22. Freeman, Rebecca, & Bonnie McElhinny (1995). Teaching language, challenging gender. In Nancy Hornberger & Sandra McKay (eds.), Sociolinguistics and language teaching. Cambridge: Cambridge University Press, 218-280. Galindo, D. Letticia (1992). Dispelling the male-only myth: Chicanas and calo. Bilingual Review 17(l):3-35. (1994). Capturing Chicana voices: An interdisciplinary approach. In Mary Bucholtz, A. C. Liang, Laurel A. Sutton, & Caitlin Hines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and. Language Conference. Berkeley, CA: Berkeley Women and Language Group, 220-231. Galindo, D. Letticia, & Maria Dolores Gonzales Velasquez (1992). A Sociolinguistic description of linguistic self-expression, innovation, and power among Chicanas in Texas and New Mexico. In Kira Hall, Mary Bucholtz, & Birch Moonwomon (eds.), Locating power: Proceedings of the Second Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 162—170. Giddens, Anthony (1990). The consequences of modernity. Cambridge: Polity Press. Gilligan, Carol (1982). In a different voice: Psychological theory and women's development. Cambridge, MA: Harvard University Press. Goffman, Erving (1961). Encounters: Two studies in the sociology of interaction. Indianapolis: Bobbs-Merrill. (1971). Relations in public: Microstudies of the public order. New York: Harper & Row. (1979). Gender advertisements. Cambridge, MA: Harvard University Press. (1981). Footing. In Goffman, Forms of talk. Philadelphia: University of Pennsylvania Press, 124-159.

Constructing Opposition within Girls' Games

407

Gonzales Velasquez, Maria Dolores (1992). The role of women in linguistic tradition and innovation in a Chicano community in New Mexico. Ph.D. diss., University of New Mexico. Goodenough, Ward H. (1981). Culture, language and society. Menlo Park, CA: Benjamin/ Cummings. Goodwin, Charles (1994). Professional vision. American Anthropologist 96(3):606-633. Goodwin, Marjorie Harness (1983). Aggravated correction and disagreement in children's conversations. Journal of Pragmatics 7:657-677. (1990). He-said-she-said: Talk as social organization among black children. Bloomington: Indiana University Press. (1998). Games of stance: Conflict and footing in hopscotch. In Susan Hoyle & Carolyn Temple Adger (eds.), Language practices of older children. New York: Oxford University Press, 23-46. Gumperz, John J. (ed.) (1982). language and social identity. Cambridge: Cambridge University Press. Gumperz, John J., & Jenny Cook-Gumperz (1982). Introduction: Language and the communication of social identity. In John J. Gumperz (ed.), Language and social identity. Cambridge: Cambridge University Press, 1-21. Henley, Nancy M. (1995). Ethnicity and gender issues in language. In Hope Landrine (ed.), Bringing cultural diversity into feminist psychology: Theory, research, practice. Washington, DC: American Psychological Association, 361-395. Heritage, John, & J. Maxwell Atkinson (1984). Introduction. In J. Maxwell Atkinson & John Heritage (eds.), Structures of social action. Cambridge: Cambridge University Press, 1-16. Hughes, Linda (1988). "But that's not really mean": Competing in a cooperative mode. Sex Roles 19:669-687. (1993). "You have to do it with style": Girls' games and girls' gaming. In Susan Tower Hollis, Linda Pershing, & M. Jane Young (eds.), Feminist theory and the study of folklore. Urbana: University of Illinois Press, 130-148. Jackson, Jean Elizabeth (1974). Language identity of the Colombian Vaupes Indians. In Richard Bauman & Joel Sherzer (eds.), Explorations in the ethnography of speaking. Cambridge: Cambridge University Press, 50-64. Jacobs, Sue-Ellen, & Jason Cromwell (1992). Visions and revisions of reality: Reflections on sex, sexuality, gender and gender variance. Journal of Homosexuality 23:43—69. Jameson, Fredric (1984). Postmodernism, or, The cultural logic of late capitalism. Durham, NC: Duke University Press. Kalcik, Susan (1975). ". . . like Ann's gynecologist or the time I was almost raped": Personal narratives in women's rap groups. In Claire R. Farrer (ed.), Women and folklore: Images and genres. Austin: University of Texas Press, 3-11. Keenan, Elinor Ochs (1974). Norm-makers, norm-breakers: Uses of speech by men and women in a Malagasy community. In Richard Bauman & Joel Sherzer (eds.), Explorations in the ethnography of speaking. Cambridge: Cambridge University Press, 125—143. Kramarae, Cheris (1990). Changing the complexion of gender in language research. In Howard Giles & W. Peter Robinson (eds.), Handbook of language and social psychology. New York: Wiley, 346-361. Kulick, Don (1992). Anger, gender, language shift and the politics of revelation in a Papua New Guinean village. Pragmatics 2(3):281-296. (1993). Speaking as a woman: Structure and gender in domestic arguments in a New Guinea village. Cultural Anthropology 8(4):5 10-541. Kyratzis, Amy (1992). Gender differences in children's persuasive justifications during pretend play. In Kira Hall, Mary Bucholtz, & Birch Moonwomon (eds.), Locating power: Proceedings of the Second Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 326—337.

408

IDENTITY AS IMPROVISATION

(1994). Tactical uses of narratives in nursery school same-sex groups. In Mary Bucholtz, A. C. Liang, Laure) A. Sutton, & Caitlin Hines (eds.), Cultural performances: Proceedings of the Third Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 389—398. Ladd, D. Robert Jr. (1980). The structure of into national meaning: Evidence from English. Bloomington: Indiana University Press. Lave, Jean, & Etienne Wenger (1991). Situated learning: Legitimate peripheral participation. Cambridge: Cambridge University Press. Le Page, Robert Brock, & Andree Tabouret-Keller (1985). Acts of identity: Creole-based approaches to language and ethnicity. Cambridge: Cambridge University Press. Leaper, Campbell (1991). Influence and involvement in children's discourse: Age, gender and partner effects. Child Development 62:797-811. Lever, Janet Rae (1978). Sex differences in the complexity of children's play and games. American Sociological Review 43:471-483. Maynard, Douglas W. (1986). The development of argumentative skills among children. In Patricia A. Adler & Peter Adler (eds.), Sociological studies of child development. Vol. 1. Greenwich, CT: JAI Press, 233-258. McElhinny, Bonnie S. (1995). Challenging hegemonic masculinities: Female and male police officers handling domestic violence. In Kira Hall & Mary Bucholtz (eds.), Gender articulated: Language and the socially constructed self. New York: Routledge, 217—244. (1998). Strategic essentialism in sociolinguistic studies of gender. In Natasha Warner, Jocelyn Ahlers, Leela Bilmes, Monica Oliver, Suzanne Wertheim, & Melinda Chen, (eds.), Gender and belief systems: Proceedings of the Fourth Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 469-480. Mendoza-Denton, Norma (1996). "Muy macha": Gender and ideology in gang-girls' discourse about makeup. Paper presented at the annual meeting of the American Association for Applied Linguistics, Symposium on Gendered Discourse Communities, Chicago. Miller, Peggy (1986). Teasing as language socialization and verbal play in a white working-class community. In Bambi Schieffelin & Elinor Ochs (eds.), Language socialization across cultures. Cambridge: Cambridge University Press, 199-212. Miller, P. M., D. L. Danaher, & D. Forbes (1986). Sex-related strategies for coping with interpersonal conflict in children aged five and seven. Developmental Psychology 22:543-548. Moore, Joan W. (1991). Going down in the barrio: Homeboys, homegirls in change. Philadelphia: Temple University Press. Moore, Sally Falk (1978). Law as process: An anthropological approach. London: Routledge & Kegan Paul. Morgan, Marcyliena (1996). Women's narratives in the English-speaking African diaspora: Identity and memory. Paper presented at the annual meeting of the American Association for Applied Linguistics, Symposium on Gendered Discourse Communities, Chicago. Ochs, Elinor (1992). Indexing gender. In Alessandro Duranti & Charles Goodwin (eds.), Rethinking context. Cambridge: Cambridge University Press, 335-358. Orenstein, Peggy (1994). Schoolgirls: Young women, self-esteem and the confidence gap. New York: Doubleday. Piaget, Jean (1965). The moral judgment of the child. New York: Free Press. (Original work published 1932) Pitcher, Evelyn Goodenough, & Lynn Hickey Schultz (1983). Boys and girls at play: The development of sex roles. New York: Praeger. Pomerantz, Anita (1984). Pursuing a response. In J. Maxwell Atkinson & John Heritage (eds.), Structures of social action. Cambridge: Cambridge University Press, 152-164.

Constructing Opposition within Girls' Games

409

Poplack, Shana (1980). "Sometimes I'll start a sentence in Spanish y termino en espanol": Toward a typology of code-switching. Linguistics 18:581-618. Rampton, Ben (1995). Language crossing and the problematisation of ethnicity and socialization. Pragmatics 5(4):485-513. Rorty, Amelie Oksenberg, & David Wong (1990). Aspects of identity and agency. In Owen Flanagan & Amelie Oksenberg Rorty (eds.), Identity, character, and morality: Essays in moral psychology. Cambridge: MIT Press, 19-36. Sacks, Harvey (1987). On the preferences for agreement and contiguity in sequences in conversation. In Graham Button & John R. E. Lee (eds.), Talk and social organisation. Clevedon, England: Multilingual Matters, 54-69. Sacks, Harvey, Emanuel A. Schegloff, & Gail Jefferson (1974). A simplest systematics for the organization of turn-taking for conversation. Language 50:696-735. Sadker, Myra, & David Sadker (1994). Failing at fairness: How America's schools cheat girls. New York: Scribner's. Schegloff, Emanuel A. (1968). Sequencing in conversational openings. American Anthropologist 70:1075-1095. Sheldon, Amy (1992). Conflict talk: Sociolinguistic challenges to self-assertion and how young girls meet them. Merrill Palmer Quarterly 38:95-117. (1993). Pickle fights: Gendered talk in preschool disputes. In Deborah Tannen (ed.), Gender and conversational interaction. Oxford: Oxford University Press, 83-109. Sutton-Smith, Brian (1979). The play of girls. In Claire B. Kopp & Martha Kirkpatrick (eds.), Becoming female. New York: Plenum, 229-257. Thorne, Barrie (1993). Gender play: Girls and boys in school. New Brunswick: Rutgers University Press. Troemel-Ploetz, Senta (1992). The construction of conversational equality by women. In Kira Hall, Mary Bucholtz, & Birch Moonwomon (eds.), Locating power: Proceedings of the Second Berkeley Women and Language Conference. Berkeley, CA: Berkeley Women and Language Group, 581-589. Turkic, Sherry (1995). Life on the screen: Identity in the age of the Internet. New York: Simon & Schuster. Turner, Victor (1985). The anthropology of performance. In Edith Turner (ed.), On the edge of the bush: Anthropology as experience. Tucson: University of Arizona Press, 177-204. (1988). The anthropology of performance. New York: PAJ. Urciuoli, Bonnie (1991). The political topography of Spanish and English: The view from a New York Puerto Rican neighborhood. American Ethnologist 18:295—310. Woolard, Kathryn A. (1997). Between friends: Gender, peer group structure and bilingualism in urban Catalonia. Language in Society 26(4):533—560. Yaeger-Dror, Malcah (1986). Intonational prominence on negatives in English. Language and Speech 28:197-230. Zentella, Ana Celia (1997). Growing up bilingual. Oxford: Blackwell.

This page intentionally left blank

NAME INDEX

Abelson, Robert, 47, 49, 55 Abrahams, Roger, 30, 31, 32, 33, 242, 252, 320 Abu-Lughod, Lila, 183 Ackerman, Lillian A., 196nl Ackroyd, Peter, 315 Adams, David Wallace, 185, 189, 196, 197, 197n2 Aftel, Mandy, 308n8 Agar, Michael, 49 Allen, Barbara, 246 Alleyne, Mervyn, 31 Allport, Gordon W., 203 Althusser, Louis, 10 Amin, Qasim, 215nl Angier, Natalie, 223 Anyon, Jean, 74, 79 Anzaldua, Gloria, 4-5, 6, 21n2, 392 Arnold, June, 334, 338-343, 340 Aronoff, Mark, 103 Aronowitz, Stanley, 355 Arrindell, Willem A., 85 Atkins, Beryl, 105 Atkinson, J. Maxwell, 97nl, 390 Atwell, Nancie, 64 Austin, Bryn, 169 Austin, J. L., 18, 343 Bahan, Ben, 62n2 Bakalti, Souad, 215nl Bakhtin, Mikhail M., 47, 110, 129, 243, 253, 318

Baldwin, Karen, 243, 246 Ball, Arnetha F., 31 Bankhead, Tallulah, 267 Barlow, David H., 87 Baron, Beth, 215nl Baron, Dennis, 333, 345n2 Barrett, Rusty, 7, 9, 10, 11, 18,

65, 70, 104, 114, 164, 234, 259, 300, 308n7, 313-327, 337, 349, 390 Earth, Fredrik, 393 Barthes, Roland, 152, 156 Bascom, William, 243 Bateson, Gregory, 84, 223, 226, 308n2, 308n4 Bateson, Mary Catherine, 238n2 Bauman, Richard, 8, 50, 242, 250, 253, 319 Baumeister, Roy, 51 Beame, Thom, 317 Bean, Judith Mattson, 8 Beaver, Patricia Dunne, 245, 251, 252 Beavin, Janet, 90 Becker, A. L., 224-225 Belenky, Mary Field, 244 Bell, Allan, 178nl, 202, 384 Bell, Genevieve, 197n2 Ben Ali, Zine El Abidine, 214 Bennett, Gillian, 243-244, 255nl Benveniste, Emile, 338

Berendt, John, 321 Bergvall, Victoria, 20, 2 1 n l , 101, 104, 123, 388 Berle, Milton, 315-316 Berman, Laine, 245 Biber, Douglas, 176 Billson, Janet Mancini, 31 Bing, Janet M., 20, 21nl, 101, 104, 123, 388 Blake, Renee, 31 Blakeslee, Sandra, 203 Blanchard, Edward B., 87 Boas, Franz, 101, 103, 106 Bodine, Anne, 5, 101, 102, 103, 117nl Bolinger, Dwight, 397 Bonvillain, Nancy, 103, 104 Booker, Karen, 108 Borker, Ruth A., 238, 274 Borland, Katherine, 245 Borman, Kathryn, 391, 392 Bourdieu, Pierre, 10, 201, 202-203, 355 Bourguiba, Habib, 211, 214 Bourke, Angela, 377, 384n4 Boxer, Andrew, 268, 269nl Boykin, Keith, 317 Bradac, James J., 403 Breuer, Josef, 84 Brice, Colleen, 117n3 Briggs, Charles L., 8, 250, 319 Brody, Jill, 372, 374 Brooke, Robert E., 64

41 1

412

Brown, Goold, 345n2 Brown, Joe, 261 Brown, Lyn M., 142n8 Browning, D. C., 156 Bruner, Jerome, 47, 85 Bryant, Dorothy, 334, 337-338 Bucholte, Mary, 3-21, 21nl, 21n3, 80n2, 104, 123, 163, 165, 238n2, 255, 317, 319, 321, 348-365, 366n4, 375, 383, 390 Buhler, Karl, 92 Bullough, Bonnie, 314 Bullough, Vern L., 314 Butler, Judith, 4-7, 21n2, 22n4, 28, 102, 108, 114, 201, 202, 213, 315, 316, 318, 327, 344 Caldas-Coulthard, Carmen Rosa, 165 Calkins, Lucy MeCormick, 64 Camarillo, Albert, 276 Cameron, Deborah, 142n3, 148, 151, 159n2, 164, 242, 255, 345nl, 349 Canfield, D. L., 278 Cantrill, H., 31, 213 Capps, Lisa, 7, 10, 13-14, 41n3, 65, 83-97, 97n4, 308n4, 391, 402 Capra, Frank, 160n8 Caputi, Jane, 154 Carranza, Isolda, 50 Castillo, Josefina M., 383 Chambless, Dianne L., 85 Cher, 314 Chcrny, Lynn, 175 Chesler, Phyllis, 85 Child, Brenda, 197n2 Christian-Smith, Linda, 65, 80n2, 348 Clinchy, Blythe McVicker, 244 Coates, Jennifer, 7, 9, 14, 2 1 n l , 42nlO, 65, 123-141,

142nlO, 146, 164, 239n8, 242, 265, 288, 308n7, 349,

380, 382, 388-389 Cobbs, Price, 31 Connell, Robert W., 73 Cook-Gumperz, Jenny, 65, 389 Cotter, Colleen, 7, 8, 19, 116, 163, 165, 181, 255, 365, 369-384, 385n9, 390 Coupland, Nikolaus, 48, 390

NAME INDEX Coyne, James, 85 Craig, Colette, 371 Creekmur, Corey K., 164 Crenshaw, Kimberle, 28 Cromwell, Jason, 390 Cronyn, Lori, 392-393 Crouch, Isabel, 6, 345nl Crowley, Matt, 267 Daly, Mary, 334 Daly, Steven, 169, 174 Danaher, D. L., 392 Daniels, Beth, 165 Davidson, Judy, 91, 282 Davies, Bronwyn, 65, 224, 225, 241, 318 Davies, Peter, 268 Davis, Angela, 325, 326 Davis, Bette, 267 De Beauvoir, Simone, 200 De Reuse, Willem, 155 Deby, Jeff, 46 Degh, Linda, 252 Deloria, Ella, 106-107, 109, 110, 112, 113 Deloria, Vine, Sr., 113 Demallie, Ray, 113 DeMarco, Joe, 317 Dennis, Everette E., 384 DiBenedetto, Tamra, 175 Dillard, J. L., 148 Diller, Barry, 366n6 DiNardo, Peter A., 87 Dobkins, Rebecca, 6, 7, 10, 15, 103, 178n4, 181-196, 371 Dobrovolsky, Michael, 103 Doi, Takeo, 238n4 Dolby, Sandra, 243 Dollard, John, 32 Dorgan, Howard, 375 Dorian, Nancy, 370, 384, 384n2 Dorrington, L. A., 192 Doty, Alexander, 164, 266, 267 Douglas, Susan J., 348 Dryman, Amy, 84, 97n3 Duberman, Martin, 316 Dubois, Betty Lou, 5-6, 345nl Dumas, Alexandre, 154 Dundes, Alan, 147, 148 Duneier, Mitchell, 30 Duranti, Alessandro, 36, 88 Dweck, Stephan, 32 Dyson, Anne Haas, 64

Eaton, William W., 84, 97n3 Eckert, Penelope, 8, 21n3, 22n4, 30, 48, 123, 142n7, 165, 201, 273, 277, 288, 317, 384, 389 Eder, Donna, 123, 405nl2 Ehrlich, Susan, 151 Eisikovits, Edina, 141n2, 142n7 Eldreth, Bessie, 245-255, 255n2, 255n3 Elgin, Suzette Haden, 343-344 Elliott, Mary, 51 Emmelkamp, Paul M. G., 85 Ervin-Tripp, Susan, 308n9 Espenson, Jane, 147 Etter-Lewis, Gwendolyn, 31 Fabrikant, Geraldine, 349 Fairclough, Norman, 142n4, 142n8 Faludi, Susan, 145, 164 Fanshel, David, 300 Farmer, John S., 153-154 Farr, Marcia, 245 Fasold, Ralph, 357 Feinberg, Leslie, 315, 316 Feldman, Carol F., 85 Ferguson, Charles, 208 Field, Margaret, 399 Finegan, Edward, 103 Fishman, Joshua, 208-209 Fishman, Pamela, 6 Flannery, Regina, 102, 103, 104 Fleisher, Julian, 314, 315, 316 Flexner, Stuart, 148 Foa, EdnaB., 84 Fodor, Iris, 85 Foley, Timothy, 375, 376 Forbes, D., 392 Foster, Michele, 31, 388, 389 Foucault, Michel, 10, 47, 85, 88, 123, 183 Frankel, Judith, 391 Franklyn, Julian, 153 Fraser, Bruce, 275-276 Freed, Alice P., 20, 2 1 n l , 104, 123, 383, 388 Freeman, Rebecca, 388 Freud, Sigmund, 84 Friedman, R. Seth, 169 Frye, Marilyn, 315 Fung, Heidi, 85 Fuqua, Vincent, 317 Fuss, Diana, 241, 344 Fusselman, Amy, 172

413

NAME INDEX Gaffney, Maureen, 375, 384n3 Gal, Susan, 65 Galindo, Letticia, 389, 392 Garber, Marjorie, 321 Garner, Thurmon, 32 Garreta, Anne, 344, 345n3 Gates, Henry Louis, Jr., 31, 32, 320 Gaudio, Rudolf P., 9, 46 Gavruseva, Lena, 229-230, 238-239n5 Geertz, Clifford, 201 Gentner, Dedre, 147 George, Nelson, 31 Gergen, Mary, 47 Gesuato, Sara, 309nl3 Gibson, Paul, 269n3 Giddens, Anthony, 390 Gilbert, Pam, 64, 65, 79 Gilbert, Sandra M., 315 Gilligan, Carol, 142n8, 244, 391-392, 402 Ginsberg, Faye, 270n6 Glass, Lillian, 321 Godel, Kurt, 343 Goffman, Erving, 9, 16, 31, 33, 36, 87, 117n8, 137, 221-224, 226, 234-237, 243, 253, 263, 296-297, 305, 390, 391, 393, 396, 404n4 Goldberger, Nancy Rule, 244 Goldstein, Alan J., 85 Gonzales Velasquez, Maria Dolores, 389 Goodenough, Ward H., 390 Goodwin, Charles, 86, 88, 90, 91, 281, 403 Goodwin, Joseph, 259 Goodwin, Marjorie, 5, 7-10, 19-20, 29, 31-35, 86, 96, 102, 123, 142n8, 196nl, 245, 274, 281, 285, 349, 377, 383, 388-404 Gould, Stephen Jay, 223 Graves, Donald, 64 Greenwood, Alice, 383 Grice, H. Paul, 17-18, 167, 294-295, 302, 308n5 Grier, William, 31 Grose, Francis, 148, 152 Gross, David M., 169 Grossman, Alan, 390 Grudin, Jonathan, 147 Gumperz, John, 207, 389

Gunderloy, Mike, 169 Gwaltney, John, 30, 31, 32, 36 Haas, Mary, 21n3, 101-105 Haddad, Tahar, 215nl Hafner, R. Julian, 85 Hall, Arsenio, 323 Hall, Deanna L., 243, 255nl Hall, Kira, 9, 21nl, 21n3, 104, 123, 165, 175, 255, 308n7, 317, 318, 321, 349 Hall, Stuart, 28, 164 Halliday, Michael A. K., 302 Hamilton, Heidi, 48 Hanks, William, 109-110, 117n8 Harding, Sandra G., 3 Harper, Phillip Brian, 317 Harre, Rom, 224, 225, 238n3, 241, 318 Harrelson, Woody, 153 Harrington, Mona, 237 Hasan, Ruqaiya, 302 Hawn, Goldie, 157 Hayes, Cassandra, 366nl0 Heath, Shirley Brice, 85, 265 Henker, Barbara, 87 Henley, Nancy M., 5, 30, 388, 403 Henley, William Ernest, 153154 Herdt, Gilbert, 268, 269n2 Heritage, John, 97nl, 390 Herring, Susan, 175 Herrmann, Claudine, 145 Herst, E. R., 84 Hertzberg, Hazel Whitman, 184 Herzog, Marvin, 201 Hex, Celina, 168 Hilbert, Jeffrey, 315 Hindley, Reg, 370, 384n5 Hines, Caitlin, 7, 10, 14-15, 21n3, 102, 145-159, 164, 197n2, 346n5 Hinton, Leanne, 159n4, 181, 185, 186, 371 Hirschberg, Julia, 283 Hobson, Dorothy, 164, 348 Hobson, Laura Z., 267 Hodge, Robert, 52 Hoffman, Dustin, 159n3 Hoffmeister, Robert, 62n2 Hofstadter, Douglas, 343 Holder, R. W., 148, 153 Holland, Dorothy, 52

Hollis, Susan, 245, 384 Hollway, Wendy, 241, 243 Holmes, Janet, 123 Hoogstra, Lisa, 85 hooks, bell, 4, 29, 244, 317 Hornyak, Janice, 232, 239n7 Horton, John, 31 Houghton, Cathryn, 73 Huffington, Adrianna, 157 Hughes, Linda, 403 Hull, Gloria T., 4 Hurston, Zora Neale, 30 Hymes, Dell, 8, 47, 319 Inoue, Miyako, 334 Irigaray, Luce, 5 Jacklin, Carol, 225 Jackson, Don, 90 Jackson, Jean Elizabeth, 389 Jackson, Raina, 31 Jacobs, Sue-Ellen, 390 Jacobson, John D., 148 Jacoby, Sally, 86 Jaffe, Alexandra, 372 James, Eli, 107, 108, 1 1 1 , 112 Jameson, Fredric, 390 Janice, Cari Goldberg, 169 Jefferson, Gail, 90, 279, 282, 287, 288, 403 Jewell, Donald P., 194 Johnson, Deborah A., 175 Johnson, Mark, 146, 154, 159nl Johnson, Sally, 21nl, 123 Johnstone, Barbara, 8, 47, 48, 101, 236, 237, 244, 252 Jones, William Allan, 187, 188, 197n5 Kalfiik, Susan, 243-244, 246, 255n1, 388 Kamler, Barbara, 64, 65 Kannapell, Barbara, 47 Kaufman, Terrence, 375 Kaye, Elizabeth, 356, 366n9 Keefe, Joan, 377 Keenan, Elinor Ochs. See Oehs (Keenan), Elinor Kelleher, Margaret, 383 Kendall, Shari, 52-53, 60, 61, 226 Kennedy, Pagan, 173-174, 176 Khosroshahi, Fatemeh, 345nl Kiberd, Declan, 372-373, 384

NAME INDEX

414

Kiesler, Sara, 51 Kiesling, Scott, 49 Kimball, Geoffrey, 103, 105, 106 King, Ruth, 151 Kinkade, M. Dale, 106 Kitzinger, Celia, 21n1 Klein, Joe, 157 Klein, Laura F., 196nl Kleiner, Liliana, 85 Kleinmann, Arthur, 266, 270n6 Kochman, Thomas, 30, 31, 32, 33 Kondo, Dorinne, 29, 31 Kramarae, Cheris, 5, 149, 343, 388 Krantz, Michael, 166 Krauss, Michael, 372 Kress, Gimther, 52 Kristeva, Julia, 264, 270n7 Krol, Ed, 175 Kroskrity, Paul, 318 Kuhn, Elisabeth, 226 Kulick, Don, 9, 389 Kunjufu, Jawanza, 31 Kurdek, Lawrence, 391, 392 Kyratzis, Amy, 67, 245, 403 Labov, William, 9, 21n3, 2930, 32, 47, 87, 200, 201, 242, 243, 248, 249, 253, 300, 317, 355, 371 Ladd, D. Robert, 397 Lakoff, George, 147, 153, 154, 159n1 Lakoff, Robin, 5, 21n3, 84, 85, 90, 105, 117n4, 146, 153, 156, 167, 177, 200, 298, 299, 300, 306, 308n5, 321, 322, 323, 324 Landers, Ann, 260, 264 Lane, Harlan, 62n2 Langellier, Kristin M., 242, 243, 244-245, 249, 255nl Largueche, Abdelhamid, 200 Largueche, Dalenda, 200 Lave, Jean, 8-9, 389 Lawler, John, 146, 155 Lawless, Elaine, 245 Le Guin, Ursula, 332, 334338, 344 Le Page, Robert, 319, 389 Leap, William, 7, 9, 10, 17, 55, 149, 164, 197n4, 238nl, 259-269, 269n2, 270n5, 270n8, 308n8

Leaper, Campbell, 392 Lee, David, 142n4 Lehrer, Adrienne, 153 Leitner, Gerhard, 383 Lensmire, Timothy, 64, 77 Lever, Janet Rae, 391, 392, 402, 403, 404n6 Levinson, Stephen, 275 Lewin, Ellen, 270n6 Lewonlin, Richard, 146-147 Liang, A. C., 7, 10, 17-18, 46, 238n1, 245, 268, 293-307 Liang, Christopher, 307nl Licari, Carmen, 281 Lighter, Jonathan Evan, 148 Lincoln, Mabel, 31 Linde, Charlotte, 47, 57, 245, 301, 308n8 Litman, Diane, 283 Livesay, Jennifer, 245 Livia, Anna, 7, 10, 18-19, 21, 21nl, 21n3, 69, 104, 159n6, 164, 255, 302, 308n7, 332345, 345n3 Lomawaima, K. Tsianina, 183, 185, 197n2 Longino, Helen E., 3 Lowe, Andrew, 164 Lucas, Jay H., 299 Lunsford, Sage, 166, 175, 176, 177 Lurie, Allison, 315 Luthin, Herbert, 105 MacAonghusa, Brian, 376 Macaulay, Monica, 117n3 Maccoby, Eleanor, 223, 225 MacGillivray, Laurie, 64 MacKenzie, Gordene Olga, 313

Madonna, 156, 314 Major, Clarence, 31, 149 Majors, Richard, 31 Malkiel, Yakov, 155 Maltz, Daniel N., 238, 274 Marcus, Eric, 316 Marks, Isaac M., 84 Marley, Bob, 4] Marshall, W. L., 85 Martin, Nomi, 31 Martinez, Ana M., 64 Marzouki, Ilhem, 213, 215nl Mason, Jeanne, 85 Matsumoto, Yoshiko, 295, 301

Maynard, Douglas W., 399 McConnell-Ginet, Sally, 8, 22n4, 30, 48, 146, 153, 165, 277, 317, 384, 389 McDonald, Mary Anne, 245 McElhinny, Bonnie, 236, 317, 366n7, 388, 389, 403 McEntire, Reba, 314 McGuire, Timothy, 51 Mcllvenny, Paul, 349 McLemorc, Cynthia, 239n7, 328n3 McLuhan, Marshall, 163 McMurray, Scott, 355 McNair-Knox, Faye, 31 Mead, Margaret, 234 Medicine, Bea, 114 Mcinhof, Ulrike, 2 1 n l , 123 Mendoza-Denton, Norma, 7, 8, 10, 17, 102, 142n8, 196nl, 236, 239n8, 273-291, 389, 392, 405nlO Mernissi, Fatima, 208 Miller, Edward K., 181, 186191, 193, 194-196 Miller, P. M., 392 Miller, Peggy J., 85, 405nl2 Mills, C. Wright, 355 Mills, Jane, 150-151, 153, 154

Mills, Kay, 376, 377 Mills, Margaret, 241 Mills, Sara, 2 1 n l , 123 Milroy, James, 375 Milroy, Lesley, 375 Mintz, Judy, 85 Mitchell-Kernan, Claudia, 21n3, 30-31, 32, 40, 320, 327 Modan, Gabriella, 273, 282 Modleski, Tania, 348 Monteiro, Kenneth P., 317 Monteria, Ivey, 32 Moonwomon, Birch, 255, 259, 269n2 Moore, Demi, 153 Moore, Henrietta L., 241 Moore, Joan W., 392 Moore, Sally, 390 Morgan, Elaine, 332 Morgan, Marcyliena, 5, 7, 10, 11, 12, 21n3, 27-41, 41n2, 96, 101, 102, 142n8, 163, 196nl, 239n7, 255, 314, 317, 320, 323, 388, 389 Morgan, Ruth, 259

415

NAME INDEX Morrison, Toni, 27 Mufwene, Salikoko, 29 Miihlhausler, Peter, 238n3 Mulac, Anthony, 403 Mumford, Laura Stempel, 348 Murdoch, Rupert, 169 Murphy, James H., 383 Murray, Lindley, 345n2 Musto, Michael, 266 Muto, Sheila, 169 Myers-Scotton, Carol, 204, 211 Nelson, Kristina, 204 Neumann, Tina, 60 Newman, Leonard, 51 Newman, Michael, 345nl Newton, Esther, 269n2, 321 Ni Chriochain, Majella, 385n8 Nichols, Patricia, 21n3 Nixon, Richard, 234 Nochimson, Martha, 348 Noel, Georgine, 344 Nussbaum, Emily, 223 Nussbaum, Jon F., 390 O'Barr, William, 105 O'Brien, Gerald T., 87 Ochs (Keenan), Elinor, 48, 84, 85, 86, 87, 88, 97n4, 108, 114, 226, 265, 295, 318, 389, 402, 404n5 O Donaill, Eamonn, 373-374 O'Donovan, Veronica, 9, 104 6 Duibhir, Padraig, 373 O'Grady, William, 103, 104 6 hAolain, Padraig, 375 Oliveira e Silva, Giselle M., 279 O Murchii, Mairtm, 370, 384nl Onifer, W., 156 Orellana, Marjorie, 5, 7, 9, 10, 13, 41n3, 57, 64-80, 141nl, 181, 245, 252, 274, 308n7, 349, 404n3 Orenstein, Peggy, 392, 404n7 O Riagain, Padraig, 370 Orwell, George, 150, 334 Padden, Carol, 59 Parker, Andrew, 8 Parks, Douglas, 113 Partridge, Eric, 150, 154 Patthey-Chavez, Genevieve, 165

Pease, Edward C., 384 Penelope (Stanley) Julia, 5, 46, 145, 146, 151, 153, 157, 346n5 Percelay, James, 32 Pershing, Linda, 245, 384 Peterson, Eric E., 245 Peterson, John L., 317 Phinney, Margaret, 64-67 Piaget, Jean, 391, 402 Picrcy, Marge, 334, 338, 341343 Pipher, Mary Bray, 142n8 Pitcher, Evelyn, 404n6 Plummer, Ken, 268-269, 269n2 Polanyi, Livia, 47, 242, 249, 253, 255nl, 255n3 Polkinghorne, Donald E., 47, 295 Pomerantz, Anita, 91, 282, 399 Poplack, Shana, 405n10 Potts, Randolph, 85 Powell, Mava Jo, 274, 279 Powers, Maria, 114 Powers, William, 107 Pratt, Richard H., 184 Preston, Dennis, 319 Prucha, Francis Paul, 185 Queen, Robin, 114 Quinn, Naomi, 52 Quirk, Randolph, 40 Quittner, Joshua, 169 Raccagni, Michelle, 215n1 Radner, Joan Newton, 245 Radway, Janice, 80n2, 348 Rampton, Ben, 142n11, 404nl Rash, Marie, 247, 249, 250 Rawson, Hugh, 148, 150 Raymond, Janice, 315 Reddy, Michael, 224 Redford, Robert, 153 Reinhardt, Robert, 263-264 Rhoads, Richard, 267, 268, 269n3 Rhodes, Richard, 146 Ribeiro, Bianca Telles, 47 Rich, Adrienne, 123, 262 Richardson, Angela, 165 Richardson, Kay, 349 Richter, Alan, 148 Rickford, John, 31

Robinson, John A., 252 Rogers, Sy, 52 Romaine, Suzanne, 142n7 Rood, David, 106 Rorty, Amelie, 404nl Rosen, Harold, 47 Ross, Diana, 314 Rostcn, Leo, 153 Rowe, Chip, 169 Rudolph, Dina, 85 RuPaul, 315-316, 323 Sacks, Harvey, 90, 282, 286, 287, 399, 403 Sadker, David, 392 Sadker, Myra, 392 Sapir, Edward, 101, 102, 103, 105, 159n6 Sarbin, Theodore, 308n8 Sass, Louis, 84 Saville-Troike, Muriel, 105, 106 Sawin, Patricia, 7, 8, 16-17, 41n3, 163, 241-255, 255n2, 265, 275, 383 Sayers, Peig, 377 Schank, Roger, 47, 49, 55 Schegloff, Emanuel, 90, 282, 286, 287, 397, 403 Schieffelin, Bambi, 84, 85, 265 Schiffrin, Deborah, 47, 61, 90, 92, 173, 255nl, 275-276, 281, 282, 379, 382 Schultz, Lynn, 404n6 Schulz, Muriel, 5, 145, 146, 150, 151, 153 Schwartz, Alan, 147 Schwenter, Scott, 280 Scollon, Ron, 238n4 Scott, James C., 183 Scott, Patricia Bell, 4 Sears, Jim, 268 Sedgwick, Eve Kosofsky, 8 Sena, Rhonda, 87 Shea, Christopher, 169 Sheldon, Amy, 403 Sherrill, Stephen, 169 Shotter, John, 47 Showalter, Elaine, 315 Shuman, Amy, 65, 383 Siegel, Jane, 51 Sigman, Marian, 87 Silverstein, Michael, 114 Simmons, Ron, 317 Simpson, O. J., 28

416

Smith, Barbara, 4 Smith, Frances, 225-226 Smith, Max C., 317 Smith, Ruth, 85 Smitherraan, Geneva, 31, 32, 320 Smolicz, Jerzy J., 374 Solsken, Judith, 65 Spears, Arthur, 29 Spears, Richard A., 148, 150 Spender, Dale, 145, 332, 346n5 Spitulnik, Debra, 372 Spivak, Gayatri Chakravorty, 263 Spock, Benjamin, 345nl Stahl, Sandra Dolby, 242 Stame, Stefania, 281 Stanley, Julia Penelope. See Penelope (Stanley), Julia Steketee, Gail, 84 Stephens, Inge, 344 Stevens, Jennifer Anne, 321, 322 Stewart, Charles, 53 Stewart, Jimmy, 160n8 Stewart, Rod, 159n3 Stoeltje, Beverly, 254 Strinati, Dominic, 348 Sutton, Laurel A., 6, 7, 10, 15, 80n2, 127, 153, 163-178, 178n5, 181, 334, 384 Sutton-Smith, Brian, 391 Swann, Joan, 142n7 Sweet, Michael J., 159n5 Sweetser, Eve, 294 Swinncy, D. A., 156 Swisher, Kara, 261 Szasz, Thomas S., 84 Tabossi, Palrizia, 156 Tabouret-Keller, Andrec, 319, 389 Talbot, Mary, 164, 165, 349 Tannen, Deborah, 7, 8, 16, 47, 50, 90, 117n8, 123, 156, 221-238, 246, 253, 255nl, 275, 299, 308n7, 362, 391, 404n5 Tarule, Jill Mattuck, 244 Tavares de Maccdo, A., 279 Taylor, Allan, 105, 106 Taylor, Anita, 20

NAME INDEX Taylor, Carolyn, 85, 86, 88 Teena, Brandon, 308n6 Thomas, Clarence, 28 Thomas, Helen, 234 Thomason, Sarah Grey, 375 Thompson, John, 202 Thompson, Sandra A., 403 Thome, Barrie, 5, 67, 274, 393 Thurston, Carol, 80n2 Tinney, James S., 317 Todasco, Ruth, 145, 148 Traugott, Elizabeth, 279 Trechter, Sara, 7, 10, 11, 1314, 21n3, 101-1 17, 181 Treichler, Paula, 149 Trevil, Gloria, 78 Troemel-Ploetz, Senta, 377, 388 Trudgill, Peter, 317 Turkle, Sherry, 390 Turner, Patricia, 34 Turner, Victor, 390, 404n4 Tyler, Liv, 149 Uchida, Aki, 201 Urban, Greg, 114, 253 Urciuoli, Bonnie, 389 Valentine, James, 159n5 Visweswaran, Kamala, 29, 31

Waddcll, Maria T., 87 Waletzky, Joshua, 87, 242, 248, 249, 253 Walkerdine, Valerie, 65, 79, 252 Walters, Keith, 5, 7, 8, 10, 1516, 22n3, 108, 117n5, 181, 200-215, 255, 308n7, 317 Ward, Martha, 265 Washington, Mary Helen, 30 Watson, Seosamh, 374 Watzlawick, Paul, 90 Weaver, Helen, 333 Wecdon, Chris, 129, 138, 139, 142n8 Wcinreich, Uriel, 201 Weir, Allison, 4 Weise, Elizabeth Reba, 175 Weiser, Ann, 297, 298, 300 Weissman, Myrna ML, 83, 84, 97n3 Wenegrat, Brant, 85

Wenger, Etienne, 8-9, 389 Wentworth, Harold, 148 Wertham, Frederic, 169, 174 West, Candace, 6 Wetzel, Patricia, 225 Whalen, Carol, 87 White, Edmund, 315 White, George, 365n2 White, Hayden, 85 White, Nancy, 31 Whittle, Stephen, 263 Whorf, Benjamin Lee, 159160n6 Wice, Nathaniel, 169, 174 Wilkes, G. A., 150, 153 Wilkinson, Sue, 21nl Williams, Donna, 234-236 Williamson, Judith, 315 Willis, Paul, 79, 164 Wilson, Robert A., 154 Winship, Janice, 348 Winterson, Jeanelte, 344 Wittig, Monique, 333, 334 Wodak, Ruth, 21 nl Wolfe, Barry E., 85 Wolfe, Susan, 46 Wolfowitz, Clare, 238n4 Wolfram, Walter, 30 Wolfson, Nessa, 92, 173 Wong, David, 404nl Wood, Kathleen, 7, 8, 10,

11, 12, 13, 41n3, 46-61, 80n2, 93, 245, 259, 27()n6,

308n8 Woods, James D., 299 Woolard, Kathryn A., 203, 317, 375, 389 Wright, Fred, 169 Yaeger-Dror, Malcah, 282, 396 Yankah, Kwesi, 31, 41n2 Yocom, Margaret, 243 Youmans, Madeleine, 165 Young, Jane, 245, 384 Young, Mark C., 84 Zamiti-Horchani, Malika, 202, 212, 213 Zentella, Ana Cclia, 389 Zimmerman, Don H., 6

SUBJECT INDEX

AADQs. See Drag queens AAE. See African American English (AAE) AAUW, 404n7 AAVE. See African American Vernacular English (AAVE) Activity-based identities, 8. See also Occasionspecific identities Adolescence. See Teenage boys; Teenage girls AE. See American English (AE) Affect. See Emotions Affiliation, 90-92, 96. See also Alignment African American English (AAE), 28, 29-30, 31, 32, 36-37, 40, 41nl, 117n5 African Americans agency and language of girls, 142n8 "behind your back/in your face" dichotomy in women's discourse, 37—41 codeswitching among, 389 cool social face and, 31-32, 36-37, 41 drag queens' performances in gay bars, 18, 313-327 exclusion and marginalization of African American women, 27-31 he-said-she-said disputes among, 33, 41 hopscotch played by children, 401 instigating in speech of, 33-36, 37, 41 intersectionality in legal decisions for African American women, 28 myth of African American rapist, 325-326 myth of promiscuity of African American women, 326—327 reading dialect and, 36-37, 40 rumors and, 33-34, 37 on shopping channel, 364, 366nl0 signifying in speech of, 30, 31, 32-33, 36, 40-41, 320, 323 speech communities of, 389 speech of girls and women, 12, 27-41, 142n8 stereotypes of, 30, 317

vernacular language as male and poor, 29-30, 317 African American Vernacular English (AAVE), 30, 317, 325 Age. See Children; Elderly Appalachian woman; Teenage boys; Teenage girls Agency and construction of identity, 65 of gay teenagers, 264-266 of teenage girls, 124, 138, 142n8 women's agency, 6, 10 Agoraphobia clinical description of, 84 definition of, 13, 84 escalating bids for responsiveness by agoraphobic, 91-92 feminist views on, 84-85 as intrapsychic disorder, 85 irrationality and, 13, 83-97 narrative interaction and agoraphobic identity, 13, 83-97 narrative roles in creation of agoraphobic identity, 88, 92-96 relationship between spouses and, 85, 90-96, 97n3 sex-role theory of, 84-85 statistics on, 84, 97n2 as "woman's syndrome," 84-85 Agreement preferences, 282, 396 Akan dry or wet speech, 41n2 Algonquian, 103 Alignment, 224, 225-226, 394-402 Alternative media, 164-178 Always Coming Home (Le Guin), 337 Ambiguity of Cooperative Principle (CP) of conversation, 297-300 WOMAN AS DESSERT metaphor and, 156 Amelioration and WOMAN AS DESSERT metaphor, 150-152 417

418

American English (AE), 36-37, 41nl American Indians. See Native Americans American Sign Language (ASL), 12, 47, 54, 56-60, 62n2, 345-346n4 Appalachian woman's stories, 16-17, 245-255 Appearance, ideologies of, 59-60 Arabic language, 203-215 Arab world. See Tunisia Arapaho, 102 Argument. See Disagreement Asian English, 132 ASL. See American Sign Language (ASL) Assimilation of Native Americans, 184-188 Atsina (Gros Venire), 102, 103, 105 Attorneys, 236, 237 Australia, 141-142n2 Authenticity on shopping channel, 358-364 Authority Lakhotalanguage and, 111-112 Native American mothers and federal boarding schools, 190-194 on shopping channel, 357-364 Autism, 234-236 Autobiography, 47, 167. See also Life stories and personal narratives Backchannel, 134, 135. See also Minimal responses Backstage talk, 137 Badness. See Goodness/badness Bagging, 32-33. See also Signifying BIA. See Bureau of Indian Affairs (BIA) Bilingualism on Irish-language radio, 374-375, 376 in Latina girls' interactions while playing hopscotch, 392-404 in Tunisia, 203-208 Biloxi, 102 Bitch as term, 178n5 Blacks. See African Americans Boarding schools for Native Americans and assimilation of Native Americans, 184188 and Christianizing Indian youth, 184 compulsory school attendance and, 185 enrollment requirements of, 185-186, 197n6 failure of, 196 health care and, 187-188 home visits of students, 186-190 mothers' attempts to circumvent school authority, 190-193 mothers' noncooperation with school authority, 193-194 Native American mothers' letters to officials at, 181-196

SUBJECT INDEX native languages prohibited at, 185 positive aspects of, 183 vocational training at, 185 Body in feminist theory, 5, 6, 108 and gender in Tunisia, 207 habitus and, 202-203 hexis and, 202-203 of host versus callers on shopping channel, 356, 366n9 in Lakhota speech-act markers, 108 in Latina/o students' stories, 70 in queer theory, 6 in sociolinguistics, 202-203 Border-crossing, 79 Border-work, 67 Boys. See Children; Latinas/os; Teenage boys Boys in the Band, 267 British teenage girls agency and, 124, 138 boys' voices in conversations of, 132 discoursal range in conversations of, 125— 129, 132-141 friendship and femininity, 137-138 gender beliefs and practices in conversations of, 14, 123-141 intertextuality in conversations of, 124, 129, 135-136 linguistic features of conversations of, 133-138 maternal voice in conversations of, 130-132 postfeminism and, 124-125 swearing by, 132 voices in conversations of, 129-136 Brothers, 260, 266 Bunny rabbit, 172 Bureau of Indian Affairs (BIA), 182, 183, 189190, 195, 197n2, 197n3 Busting, 32-33. See also Signifying California adolescent Latinas in Northern California, 273 California Style Collective, 274 Greenville Indian School, 182-196 Latina children in Pico Union, 401 Native American languages, 371 Capping, 32-33. See also Signifying Category model of identity, 7-8, 318 Category of gender, 389-390 Children "aggravated correction" in arguments of, 396 assertiveness of working-class white children, 405nl2 boarding schools for Native American children, 182-196

SUBJECT INDEX conflict avoidance of middle-class children, 405 gendered identities of, 391-392 health care of Native American children, 187-188 hopscotch played by, 19-20, 392-404 moral development of, 391-392, 402 stories of Latina and Latino primary-school children, 13, 64-80 and storytelling during family dinners, 86 See also Teenage boys; Teenage girls China, 228 Chorus Line, 261 Chukchee, 102 Class assertiveness of working-class white children, 405nl2 conflict avoidance of middle-class children, 405 resistance of working-class teenagers to regulatory forces of school, 79 shopping channel and, 354-357, 364, 365nl, 366n6 and stories of Latina/o primary-grade students, 65-80 teenage girls' speech and, 124, 142n7 turn-initial no among Latina adolescents, 273-291 Closet language of, 267-269 See also Coming out; Gay implicature; Gay men; Lesbians CMC. See Computer-mediated communication (CMC) Coconstruction, 35, 36, 92-93, 390 Codeswitching, 6, 204, 208-209, 389, 405nl0 Cognitive linguistics, 145 Coherence, 47, 167 Collaborative narratives, 243 Collaborative opposition among Latina adolescents, 273-291 Colonialism internal colonialism, 370, 384n2 postcolonial Tunisia, 203-215 Coming out Deaf and hearing lesbians' stories of, 12, 46-61 frames in lesbians' stories of, 50-52 function of, 295 gay male experiences of, 17, 259-269 heterosexist ideologies in lesbians' stories of, 52-53, 57, 60 Common-gender pronoun systems. See Epicene (common-gender) pronoun systems Communication. See Conversation; Interaction; Language

419 Community-of-practice model, 8-9, 389 "Compulsory heterosexuality," 123, 142n3 Computer-mediated communication (CMC), 54-58, 167-168, 174-178 Conflict. See Disagreement Conflicting identity, 317 Conflict management in Latina teenagers' conversation, 17, 273-291 Confrontation by African American teenage girls, 35 Connection and status, 227-234 Consciousness raising (CR), 14, 126-128, 135138 Consenting Adults, 260-261, 267 Consumerism and shopping channel, 19, 348-365 Context in Lakhota language, 110-114 of women's narrative, 248-250, 252-253 Conversation among African American women and girls, 30-41 agoraphobic identity and family interactions, 86-97 of British teenage girls, 125-141 conversational devices, 297-300 conversational maxims, 167-168, 294-295, 298-299, 300, 303, 305, 306, 308n5 conversational stratagems, 297-300 Cooperative Principle (CP) of, 293-307 gay implicature, 300-307, 308nn9-10 Grice's conversational logic, 294-295, 308n5 between men, 229-232 turn-initial no among Latina teenagers, 273291 public versus private sphere, 129, 369, 370, 372, 375, 376, 380, 382, 388-389 in workplace, 16, 221-238 See also Discourse; Discourse markers (DMs); Interaction; Language Conversational signifying, 30, 31, 32-33, 36, 40-41, 320, 323 Cook and the Carpenter (Arnold), 334, 338-341 Cool social face, 31-32, 36-37, 41 Cooperation by middle-class women and girls, 388, 389, 402, 405 Cooperative Principle (CP) of conversation ambiguity of, 297-300 and conversational devices and stratagems, 297-300 gay implicatures and, 300-307, 308nn9-10 and gender as basic implicature, 295 and genderless reference terms, 301-302 Grice's conversational logic, 294-295, 308n5 heterosexual presumption and, 295 lesbian/gay identity and, 293-307

420

Cooperative Principle (CP) of conversation {continued} and managing lesbian/gay identity, 296-297 and problem of lesbian/gay self-presentation, 295-296 Correspondence. See Letters to boarding-school officials CP. See Cooperative Principle (CP) of conversation CR. See Consciousness raising (CR) Cross-cultural variation in children's games, 401-402 Cross-dressers, 314 Cross-sex interactions, 6, 7, 15-16, 67, 69, 206215, 388 Culture cross-cultural variation in children's games, 401-402 culturally approved identities, 49, 293-294, 306-307 Deaf culture, 6 l-62n2 gay culture, 259, 269n2 identity as "operating culture," 390 languaculture, 49 new cultural forms, 349 queer culture, 46, 61nl, 114, 263 shopping channel and, 349-350 in Tunisia, 15-16, 200-215 Daughters Inc., 341 "Dead metaphors," 157 Deal' as born into hearing families, 60 coming-out stories of Deaf lesbians, 12, 5461 and "Deaf-World," 61-62n2 identity of, 60-61 Deathtrap, 261 Department stores, 355, 356 Desired woman metaphor, 146 DESSERT metaphor. See WOMAN AS DESSERT metaphor Dialect reading dialect by African Americans, 36-37, 40 See also Vernacular Difference as theory of language and gender, 101 See also Gender differences Diglossia, 208-209 Disagreement aggravated correction, 396 in British teenage girls' conversation, 135 he-said-she-said disputes, 33, 41

SUBJECT INDEX

among Jewish Americans, 282 in Native American mothers' letters to boarding school officials, 188-196 in playing hopscotch, 397-401, 403 turn-initial no among Latina adolescents, 273-291 Discourse Anglo discourse structure on Irish-language radio, 374-375, 376 of British teenage girls, 125-141 of computer-mediated communication (CMC), 168 definition of, 142n4 of gender segregation in stories by Latina/o students, 70, 79 of shopping channel, 350-365 workplace discourse, 16, 221-238 of zines, 168 See also Conversation; Discourse markers (DMs); Language Discourse analysis, 4, 349, 365 Discourse markers (DMs) definition of, 275 English discourse markers on Irish-language radio, 377-382 Latina teenagers' use of no as, 273-291 social meaning and, 275-276 Displays, 223-224, 234, 391 DMs. See Discourse markers (DMs) Dominance, as theory of language and gender, 5-6 Dozens, 32-33. See also Signifying Drag kings, 314 Drag queens African American drag queens, 18, 313-327 compared with other transgender groups, 313-314, 318 definition of, 314 feminist views of, 315 in fiction, 321 glam queens, 314 language of, 317-327 "passing" by, 314, 319 performed identity and, 317-320, 323-327 profanity and vulgarity used by, 324-325 as professional performers, 314—315 queer theorists on, 315, 316 signifying and, 320 as subversive versus misogynistic, 315-317, 325-327 white women's language among African American drag queens, 321-327

SUBJECT INDEX Earthsong (Elgin), 343-344 Education compulsory Irish in Irish schools, 384n5 of Native Americans, 182-196 resistance of working-class teenagers to regulatory forces of school, 79 stories of Latina and Latino primary-school children, 13, 64-80 in Tunisia, 204-206, 214 E-mail. See Internet Embodiment. See Body Emotions of agoraphobic, 89-90 British teenage girls' talk about, 138-139 of gay teenagers, 259-269 in lesbian coming-out stories, 54-61 narratives on, 85-86, 89-90 in playing hopscotch, 397-399 Endangered languages Irish language, 19, 370-371, 374-376 Native American languages, 103, 181 and reduction of gendered forms, 105-106 See also Language preservation and revival English. See African American English (AAE); African American Vernacular English (AAVE); Standard English Epicene (common-gender) pronoun systems, 18-19, 332-345 Ethnicity and race. Sec African Americans; European Americans; Jews; Latinas/os; Native Americans; Women of color Ethnography, 18, 28, 30, 66, 276, 348, 369, 403 Eunuchs (hijras), 104-105 European Americans African American drag queens' use of white women's language, 321-327 attitudes toward African Americans' speech, 364 as gay men, 260-262 Lakhota's imitation of, 111 Native Americans' letters to boarding-school officials, 15, 181-196 white middle-class heterosexuals as research norm, 4, 9, 11, 65, 241, 244, 388 stereotypes of white middle-class women, 321-322 as speakers of Standard English, 29 Eace "behind your back/in your face" dichotomy in African American discourse, 37-41 cool social face and African Americans, 3132, 36-37, 41 definition of social face, 31 Goffman on, 224

421

"making face" as metaphor for identity, 5 marker now used in face-saving repair, 379 Family interaction agoraphobic identity and, 86-97 storytelling during family dinners, 86, 88 Fathers during family dinners, 86, 88 Feelings. See Emotions Female impersonators, 314, 321. See also Drag queens Femininity British teenage girls and, 14, 123-141 and discursive struggle, 139 friendship and, 137-138 ideologies of, 85, 123-141 in stories by Latina/o primary-grade students, 67, 70-79 See also Gender identity; WOMAN AS DESSERT metaphor; WOMAN AS OBJECT metaphor Feminism and British teenage girls' conversations, 126, 127, 128 cultural feminism, 388-389 Islamic feminism, 208, 211-215 multicultural feminism, 6 and multiple identities, 4-7, 20, 28 postfeminism and, 124-125 and pronoun politics, 18-19, 21, 332, 344345 radical feminism, 21, 145 women of color and, 4-7, 29 Feminist theory on agoraphobia, 84-85 on biocultural presuppositions underlying gender oppositions, 101 on bodies, 5, 6 and complexities of women's identities, 6 on drag queens, 315 Hall on, 28 on insanity or irrationality, 84-85 on language, 151-152 language and identity crisis in, 3-7 on pleasure, 365 on popular culture, 364-365 on pronouns, 21, 332, 344-345 on relationship between word and world, 151-152 women of color and, 4-7, 29 on women's language, 84 Fiction drag queen in, 321 gay male characters in, 263-264 Le Guin's use of masculine pronoun, 334337 nongendcred pronoun systems in feminist fiction, 18-19, 332-345

422

Folklore, 30, 110, 369, 377 Food terms for men, 155, 159nn2-3 WOMAN AS DESSERT metaphor, 14-15, 145-158 Footing, 391 Frame and framing Goffman on, 221 Hanks's concept of, 109, 117n8 in Lakhota language, 109-110 in narrative discourse, 50-52, 54-58, 59, 60 refraining of gender roles through media language, 376-384 reframing of shopping channel discourse, 356-364 sex-class-linked framing, 16, 222-227, 229238 workplace discourse and, 221-238 Framework in Lakhota language, 109-110 French language epicene (common-gender) pronoun in, 333, 345n3 in Tunisia, 203-205, 207-208 Friendship among British teenage girls, 137-138 in stories by Latina/o primary-grade students, 67,69 Gaelic, 384nnl-2 Gaeltacht, 370, 372-375 Games African American children's language games, 32-36 of instigating among African Americans, 3336, 37, 41 Latina girls' interactional practices while playing hopscotch, 19-20, 388-404 of signifying among African Americans, 32-33 Gangs, 29-30, 236 Gay culture, 259, 269n2 Gaydar, 266-267 Gay implicature, 300-307, 308nn9-10 Gay men African American drag queens, 18, 313-327 coming-out experiences of gay teenagers, 17, 259-269, 270n5 conversational devices and stratagems used by, 297-300 and Cooperative Principle of conversation, 17-18, 293-307 covert meanings in communication of, 18, 300-304 gay implicature and, 300-307, 308nn9-10 heteronormativity versus gay imaginary, 262264 homophobia against, 267-268

SUBJECT INDEX ideology of, 60 and language of the closet, 267-269 media and, 46, 260-261, 264, 266-267, 315316

misogyny by, 316 name-calling by peers, 261, 263 Native Americans as, 115, 117n6 "passing" by, 297, 305, 307 self-managed socialization by gay teenagers, 264-266, 268-269 self-presentation by, 295-296, 307 suicide of gay teenagers, 269n3 text-making and, 259-260 and use of feminine reference terms, 238nl Gender dichotomy of, 101-104, 201, 241, 388-389, 391-392 Goffman on, 222-224 Maccoby on, 223 as master category, 389-390 sex versus, 223 Gender-bending, 390 Gender differences in attorneys' styles, 237 in individual abilities, 225 in moral development, 391-392, 402 in narrative, 241-244 in sermons, 225-226 in smiling, 236-237 in university professors' styles, 226 public versus private sphere, 129, 369, 370, 372, 375, 376, 380, 382, 388-389 in workplace discourse, 229-234 Gender-exclusive languages, 103-106 Gender identity agoraphobia and, 84-85 in British teenage girls' conversations, 14, 123-141 of children generally, 391-392 of drag queens, 313-327 heterosexual presumption and, 295 on Internet, 390 Irish-language radio and, 376-384 in Lakhota language, 13-14, 101-117 literacy and, 65 lived experience and theories of, 200-202 models of women's narrative, 16-17, 241-255 in Native American languages generally, 102-106 politics of, 10-11 popular culture and, 348-350 power and, 201 shopping channel and, 19, 348-365 in stories of Latina and Latino primary-school children, 13, 64-80

SUBJECT INDEX of Tunisians, 15-16, 200-215 workplace discourse, 16, 221-238 Generic masculine pronoun, 5, 332, 334-337 Genre features in alternative media, 170-176 Genre frame, 51 Germany, 226 Girls. See Children; Latinas/os; Teenage girls Goodness/badness in elderly Appalachian woman's storytelling, 252 "good-girl research," 9, 11 in stories by Latina/o primary-grade students, 73-77, 78, 79 Grammar of Lakhota, 102-117 Gros Ventre (Atsina), 102, 103, 105 Habitus, 202-203 Hate speech, 102 Hawaii, 371 Hedging, 112-113, 135, 136 He-said-she-said disputes, 33, 41 Heteronormativity definition of, 262 versus gay imaginary, 262-264 Heterosexism "compulsory heterosexuality," 123, 142n3 in lesbian coming-out stories, 52, 57, 60 media portrayal of, 52-53 See also Homophobia Heterosexual presumption, 295 Hexis, 202-203 Hierarchy in office interactions, 16, 221-238 Hijeb, 214 Hijras (male eunuchs), 104-105 Hip hop, 42n9 Hispanics. See Latinas/os History of Shadows (Reinhardt), 263-264 Hokan, 103 Home Shopping Network, 364, 365nl, 366nlO. See also Shopping channel Homophobia, 267-268, 327. See also Heterosexism Homosexuals. See Coming out; Gay men; Lesbians; Queer Hopscotch boys playing, 405n9 breaking the rules, 399-401 cross-cultural variation in, 401-402 fieldwork on, 392-393 Lalina girls' interactional practices while playing, 19-20, 392-404, 405n8 occasion-specific identities within, 393-395, 396 out calls in, 395-397, 398, 401

423

participant structure and sequential structure of out-call sequences in, 395-396 Pico Union children and, 401 pitch contours in, 396-397 turn construction in opposition moves in, 397-399 Identity acceptable selves, 48, 49, 50, 52, 55, 61 of African American drag queens, 18, 313-327 of African American women, 28-29 agoraphobic identity, 13, 83-97 in alternative publishing, 163-178 Anzaldua on, 5, 6, 7 in British teenage girls' conversations, 14, 123-141 Butler on, 5, 7, 28, 213 category model of, 7-8 children's gendered identities, 391-392 community-of-practice model of, 8-9 conflicting identity, 317 Cooperative Principle (CP) and lesbian/gay identity, 293-307 culturally approved identities, 293-294, 306-307 Deaf identity, 60-61 definition of, 404nl "failed" identity, 317 feminist theory generally and, 4-7 of fraternity men, 49 and gay teenagers' coming-out experiences, 17, 259-269 Goffman on, 222-223 imposed, 294 Irish-language radio and, 19, 369-384 and language practice generally, 48, 64-65, 389-391 of Latina and Latino student writers, 64—80 and lesbians' coming-out stories, 46-61 in life stories and personal narratives, 46-47, 242-243 "making face" as metaphor for, 5 models of women's narrative, 16-17, 241-255 occasion-specific identities, 8, 391, 393-395, 396 performed identity, 8, 307, 317-320, 323-327 politics of gender identity, 10—11 popular media and, 164 repertoire of identity, 318, 390 shopping channel and, 19, 348-365 sociolinguistic approach to, 317 transgressive identities generally, 9-10 of Tunisians, 15-16, 200-215 workplace discourse, 16, 221-238 See also Gender identity; and headings beginning with Self

424

Ideologies of appearance, 59-60 definition of, 52 of femininity, 85, 123-141 of gender in British teenage girls' conversations, 14, 123-141 heterosexist ideologies, 52-53, 57, 60 of homosexuality, 60 in lesbians' coming-out stories, 52-53, 55, 57, 58, 60 and letters of Native American mothers to boarding-school officials, 15, 181-196 overview of identity as, 14-16 patriarchal ideologies, 348 in Tunisian gender and culture negotiation, 15-16, 200-215 WOMAN AS DESSERT metaphor, 14-15, 145-158 iGuide, 166 Illiteracy. Sec Education; Literacy Implicature gay implicature, 300-307, 308nn9-10 gender as, 295 Indecent Proposal, 153 Indians. See Native Americans Instigating, 33-36, 37, 41. See also He-saidshe-said disputes Interaction among African American women and girls, 30-41 in African American children's language play, 32-36 agoraphobic identity and family interactions, 86-97 "behind your back/in your face" dichotomy in, 37-41 of British teenage girls, 125-141 cool social face and, 31-32, 36-37, 41 cross-sex interactions, 6, 7, 15—16, 67, 69, 206-215, 388 emergent narratives as interactional, 86 gay implicature, 300-307, 308nn9-l() he-said-she-said disputes, 33, 41 immanent reference of talk, 300 Latinas" interactional practices while playing hopscotch, 19-20, 388-404 in lesbians' coming-out stories, 48-49 Native American women's letters to boarding school officials, 181-196 reading dialect, 36-37 resources for and constraints on, 48-49 between shopping channel host and callers, 357-364 turn-initial no among Latina adolescents, 273-291

SUBJECT INDEX workplace interactions, 16, 229-238 See also Conversation; Discourse; Language Internal colonialism, 370, 384n2 Internet e-mail, 12, 51, 54-58, 308-309n11 gender blurring on, 390 online journals, 166-168, 174-178 online shopping, 350, 365n2 Intertextuality in British teenage girls' conversations, 124, 129, 135-136 definition of, 48 resources and constraints and, 48-49 Intimacy. See Friendship; Love and romance Ireland compulsory Irish in schools in, 384n5 and Irish language generally, 370-371, 384nl, 385n7 preservation ideology in, 376-377, 384n4 women and language maintenance, 371 women's role in, 375-384, 384n3, 385n8 See also Irish-language radio Irish language, 370-371, 374-376, 384nl. See also Irish-language radio Irish-language radio Anglo discourse structure on, 374-375, 376 effects of, on language maintenance and identity, 374-376 English discourse markers on, 377-382 and enhancement of status of language, 372 Raidio na Gaeltachta (RnaG), 372-376, 379 Raidt'o na Life (RnaL), 373-382, 384 reframing of gender roles through media language, 376-384 and use of Irish language generally, 370-371, 375, 384nl women's participation in, 375-384, 384n3, 385n8 Irish-language television, 376 Irrationality and agoraphobia, 13, 83-97 Islam, 15-16, 200-215 Italy, 309nl3 It's a Wonderful Life, 160n8 Japan, 225, 228, 295 Japanese language, 108, 114 Java, 228 Jellybean, 177 Jews collaborative disagreement among Jewish Americans, 282 definition of Jewishness, 308n3 Joning, 32-33. See also Signifying Journals. See Online journals Judas Rose (Elgin), 343-344

SUBJECT INDEX "Kernel stories," 243, 255n1 Kin of Ata Are Waiting for You (Bryant), 334, 337-338 Koasati, 102-106 Lakhota language appropriations in, 115-116 authoritative speech, 111-112 decline of, 181 Deloria texts of, 106-107 fieldwork context of, 107 frame and framework in, 109-110 gender in, 107-110, 112 hedging and diffidence expressed in, 112-113 joking in, 113 native speakers on, 116 nonstereotypic meanings in, 114-116 number of speakers of, 106 quotation in, 114—115 sentence-final particles in, 107-108 stereotypic meanings in context, 110-14 "women's language" and "men's language" in, 13-14, 101-117 LA. Law, 355 Lames, 29-30, 317 Language of African American drag queens, 18, 317327 African American English (AAE), 28, 29-30, 31, 32, 36-37, 40, 41nl of African American girls and women, 12, 27-41, 142n8 agoraphobic identity and, 83-97 body and, 203 diglossia, 208-209 epicene (common-gender) pronoun systems, 18-19 feminist views on, 4-7, 151-152 gay implicature, 300-307, 308nn9-10 and gay teenagers' coming-out experiences, 17, 259-269 gender beliefs and practices in British teenage girls' conversations, 14, 123-141 gendered language of Native Americans generally, 102-106 gender-exclusive languages, 103-106 hate speech, 102 inventing identities through, 64-65 Irish-language radio and, 19, 369-384 Lakhota language as gendered, 13-14, 101-117 as languaging, 224-225 masculine pronoun as generic, 5, 332, 334337 no as discourse marker by Latina teenagers, 273-291

425

overgeneralized male forms, 5 pejorative referring terms, 5 performance and, 8 popular culture and, 15, 348-350 queer languages, 114 relationship of identity to generally, 48, 389391 of restraint, 267-268, 270n8 Sapir-Whorf hypothesis of, 159-160n6 in Tunisia, 203-215 vernacular versus standard English, 29-30 white women's language used by African American drag queens, 321-327 WOMAN AS DESSERT metaphor, 14-15, 145-158 women and maintenance of, 371 Language games. See Games Language maintenance, 371, 374-376 Language preservation and revival, 181, 371, 376-377, 384n4 Latinas/os agency and language of Latinas, 142n8, 274, 392, 403 alignment and disalignment in discourse of teenage girls, 17, 273-291 interactional practices of girls while playing hopscotch, 19-20, 388-404 self-esteem of Latina teenagers, 404n7 sexuality and reproductive choices of Latinas, 73 stories of primary-school children, 13, 64-80 Left Hand of Darkness (Le Guin), 334-337 Lesbians and appearance ideologies, 59-60 "baby dykes," 55 butch lesbians, 49, 59-60 coming-out stories of, 12, 46-61 conversational devices and stratagems used by, 297-300 and Cooperative Principle of conversation, 17-18, 293-307 covert meanings in communication of, 18, 300-304 gay implicatures and, 300-307, 308nn9-10 genderless reference terms used by, 301-302 heterosexist ideologies in coming-out stories of, 52-53, 57, 60 ideology of, 58, 60 as mothers, 270n6 Native Americans as, 115-116 "passing" by, 297, 305, 307 self-presentation by, 53-61, 295-296, 304, 307 text-making and, 259-260 Letters to boarding-school officials, 15, 181196

426

Life stories and personal narratives coherence in, 47 of elderly Appalachian woman, 16-17, 245-255 features of, 242 frames in, 50-52 of gay men, 260-269, 269n2, 270n5 identity in generally, 46-47 of lesbian mothers, 270n6 lesbians' coming-out stories, 12, 46-61 models of women's narrative, 241-255 "personal-experience narrative" and identity, 242-243 subjective nature of, 270n6 See also Narrative Linguistic anthropology, 389, 402 Linguistics. See Cognitive linguistics; Psycholinguistics; Sociolinguistics Literacy gender and, 65, 69, 182, 214 inventing identities through, 64-65 See also Education Love and romance in British teenage girls' conversations, 126, 127 in stories by Latina/o primary-grade students, 71-74 Malagasy, 389 Marginality, 8-9, 27-31, 164, 165 Masculine pronoun as generic, 5, 332, 334-337 Masculinity and discursive struggle, 139 as normative, 11 in stories by Latina/o primary-grade students, 74-77 See also Gay men; Men's language; and headings beginning with Gender Mass media. See Popular media Maternal role. See Mothers Meaning covert meanings in lesbian and gay communication, 18, 300-304 discourse markers (DMs) and, 275-276 in Lakhota language, 110-116 Media. See Alternative media; Popular media Men. See Gay men; and headings beginning with Gender Men's language in cross-sex interactions, 6, 7, 15-16, 67, 69, 206-215, 388 fathers during family dinners, 86, 88 in Lakhota, 13-14, 101-117 in public sphere, 388-389 in Tunisia, 206-215 See also Gay men; and headings beginning with Gender Menstruation, 135-141

SUBJECT INDEX Metaphors "dead metaphors," 157 desired woman metaphor, 146 "making face" as metaphor for identity, 5 PEOPLE AS OBJECTS metaphors, 147 power of, 146-147 WOMAN AS DESSERT metaphor, 14-15, 145-158 WOMAN AS OBJECT metaphors, 14, 146 Midnight in the Garden of Good and Evil (Berendt), 321 Minimal responses, 136, 137 Minorities. See African Americans; Latinas/os; Native Americans; Women of color Mirroring, 126, 142nlO Misogyny, 315-317, 325-327 Mixed-sex groups. See Cross-sex interactions Moral development, 391-392, 402 Morphology, 102-103 Mothers language maintenance by, 371 letters of Native American mothers to boarding-school officials, 15, 181-196 maternal voice in British teenage girls' conversation, 130-132 and storytelling during family dinners, 86 Muskogean, 103, 108 Narrative agoraphobic identity and, 85-97 conarrators and, 86 construction of irrationality in, 89-97 definition of, 87 emergent narratives as interactional, 86 on emotions, 85-86 during family dinners, 86, 88 family history of, 93 function of, 65 of gay men, 259-269 gender differences in, 241-244 involvement of interlocutors in, 92 narrative roles, 86, 88, 92-96 See also Life stories and personal narratives; Women's narrative; and headings beginning with Stories Narrative prototypes, 49, 51 Narrative skeletons, 49, 51, 55-57, 59-61 Nationalism, 209, 389. See also specific countries Native Americans assimilation of, 184-188 boarding schools for, 182-196 English spoken and written by, 197n4 homosexuality and, 115-116, 117n6 gendered language of generally, 102-106 health of and health care for, 187-188

SUBJECT INDEX Lakhota "women's language" and "men's language," 13-14, 101-117 languages of, in decline, 103, 105-106, 116, 181, 371 literacy levels of, from 1916-1922, 182 Native Tongue (Elgin), 343-344 Navajo, 104 Neologisms, 334-344 New Guinea, 389 Nonstandard language in alternative media, 165, 168, 170-172, 176 by shopping channel callers, 356-357 See also African American English; African American Vernacular English; Vernacular Novelas on Spanish television, 73, 74, 80n2 Novels. See Fiction Objectification. See WOMAN AS DESSERT metaphor; WOMAN AS OBJECT metaphors Occasion-specific identities, 391, 393-395, 396 Oklahoma, 102, 183 Online journals, 166-168, 174-178 Opoponax (Wittig), 333, 345n3 Opposition. See Disagreement; Resistance Other American Dialects (OAD), 30 "Passing" by drag queens, 314, 319 by female impersonators, 314 by lesbians and gay men, 297, 305, 307 Patriarchal ideologies in British teenage girls' conversations, 127128, 135, 136-137, 138 popular culture and, 52-53, 163, 348 Pejoration, 150-152 Performance, anthropology of, 390 Performance frame, 50, 59 Performances of drag queens, 18, 313-327 Performed identity, 8, 307, 317-320, 323-327 Personal narratives. See Life stories and personal narratives Phonetics, 152, 154-156 Phonology, 102-103, 209-211 Phonosemantics, 155-156 Pitch contours in hopscotch participants, 396-397 Pixxiebitch, 164, 171, 178n5 Play. See Games Pleasure, 348-349, 350, 352, 363-365 Polysemy, 222, 320 Popular media alternative media, 15, 163-178 computer-mediated communication (CMC), 167-168, 174-178 control of "mainstream" media, 165 discourse analysis and, 349

427

and enhancement of status of language, 372 heterosexist ideologies portrayed by, 52-53 homosexuality portrayed in, 46, 260-261 identity and, 164 Irish-language radio, 19, 369-384 Irish-language television, 376 newspaper and magazine articles on teleshopping, 356, 366n8 Novelas on Spanish television, 73, 74, 80n2 online journals, 166-168, 174-178 patriarchal ideologies in, 348 retrieving gay messages from, 264, 266-267 self-publishing and, 164-178 sexism and, 52-53, 163, 348 shopping channel, 19, 348-365 soap operas, 125, 130, 267 sociolinguistics and, 163, 164-165 targeted to women, 163, 351 Tunisian television, 214-15 women's participation in, 375-384, 384n3, 385n8 zines, 165-174, 177-178 Positioning, 224, 225 Postcolonialism, 203-215 Postfeminism, 124-125 Postmodernism, 28 Poststructuralism, 241 Poverty, 29-30, 77-78, 317 Power American view of, 225 Cooperative Principle of conversation and, 17-18, 293-307 dominance as theory of language and gender, 5-6 Japanese view of, 225 Native American women's letters to boarding school officials, 181-196 politics of gender identity, 10-11 power imbalances in cross-sex interactions, 6 resistance and, 183 in stories by Latina/o primary-grade students, 77-78 See also Status Practice Butler on, 22n4 definitions of, 22n4 Eckert and McConnell-Ginet on, 8-9, 22n4 identity and practice-based analysis, 8-9, 317, 389 "marginal" members and, 8-9 Pragmatics Cooperative Principle of conversation and, 17-18, 293-307 covert meanings in lesbian and gay communication, 18, 300-304 Latina teenagers' alignment and disalignment discourse, 17, 273-291

428

Presentation of self. See Self-presentation Primary Colors, 157 Profanity and vulgarity in British teenage girls' conversations, 132 by drag queens, 324-325 in workplace discourse by men, 230 Promiscuity myth of promiscuity of African American women, 326-327 in WOMAN AS DESSERT metaphor, 150-154 Pronouns Baron on failure of nongendered pronoun systems, 333-334 demonstrative pronoun that in African American English, 40 epicene (common-gender) pronoun systems in feminist fiction, 18-19, 332-345 examples of nongendered pronoun systems, 333 gay men's use of feminine reference terms, 238nl masculine pronoun as generic, 5, 332, 334—337 one used instead of generic he, 332-333, 345n3 radical-feminist linguistic activities on, 21 she and her used as generic pronouns, 345nl she and he used alternately in texts, 345nl singular they, 332, 345n2 Prototypes. See Narrative prototypes Psycholinguistics, 156 Psychopathology, 83-96 Publishing alternative publishing, 15, 163-178 online journals, 166-168, 174-178 zines, 165-174, 177-178 Queen of the Thundercats, 163 Queer definition of, 61nl heteronormativity and, 263 language, 114 media portrayal of queers, 46 See also Coming out; Gay men; Lesbians Queer theory, 6, 315, 316 Questions in British teenage girls' conversation, 134 QVC. See Shopping channel Race. See African Americans; Ethnicity and race; European Americans; Latinas/os; Native Americans; Women of color Racism, 6, 27-31, 317, 390 Radio. See Irish-language radio; Popular media Raidio na Gaeltachta (RnaG), 372-376, 379 Raidio na Life (RnaL), 373-382, 384 Ranking, 32-33. See also Signifying

SUBJECT INDEX Rape, 325-326 Rapport talk, 232-233 Reading dialect, 36-37, 40 Reference, 300-302 Regional accents, 356-357 Repertoire of identity, 318, 390 Resistance identities and, 96 low-affiliative responses to agoraphobic, 9092, 96 in Native American mothers' letters to boarding school officials, 188-196 power and, 183 and white women's language used by African American drag queens, 321-327 RnaG. See Raidio na Gaeltachta (RnaG) RnaL. See Raidio na Life (RnaL) Role gender roles in Irish-language radio, 376-384 narrative roles in storytelling, 88, 92-96 sex-role theory, 84-85 teacher's role in writing workshop, 67, 72-73 women's role in Ireland, 375-384, 384n3, 385n8 women's role in Tunisia, 208, 211-215 Romance. See Love and romance Rumors, 33-34, 37 Sapir-Whorf hypothesis, 19, 159-160n6, 336, 343-344 Schools. See Boarding schools for Native Americans; Education Science fiction. See Fiction Self. See Identity; and headings beginning with Self Self-disclosure in British teenage girls' conversations, 14, 126-128, 135-138 in lesbian coming-out stories, 51 Self-esteem, 404n7 "Self-hate" of women, 348 Self-managed socialization by gay teenagers, 264-266, 268-269 Self-presentation of elderly Appalachian woman, 252-253 lesbian/gay self-presentation, 295-296, 304, 307 of speaker versus addressee, 307-308n2 Self-transformation, 46-47. See also Coming-out Semantics phonosemantics, 155-156 and WOMAN AS DESSERT metaphor, 150-156 Sex-class-linked framing, 16, 222-227, 229238 Sex differences. See Gender differences

SUBJECT INDEX Sexism. See Misogyny; Patriarchal ideologies; WOMAN AS DESSERT metaphor Sexual orientation. See Coming out; Drag queens; Gay men; Heteronormativity; Heterosexism; Heterosexual presumption; Lesbians; Queer Sex versus gender, 223 Shopping channel African Americans on, 364, 366n10 audience of, 351, 366n4, 366n6 callers' testimonials on, 358-364, 366n5 class arrangements on, 354-356, 364, 365nl, 366n6 competing construction of identity on, 362364 description of, 349-350 host's linguistic practices on, 357-364 men on, 366n4 nonstandard forms of English used by callers, 356-357 pleasure associated with, 350, 352, 363-364 refraining of discourse on, 356-364 regional accents of callers, 356-357 Signifying in African American children's games, 32-33 by African American drag queens, 320, 323 in African American women's speech, 30, 31, 36, 40-41 definition of, 32 Silence, 263, 268 Siouan, 102, 103, 106 Situated activity system, 391 Slang. See Profanity and vulgarity; WOMAN AS DESSERT metaphor Smiling, 235-237 Snapping, 32-33. See also Signifying Soap operas, 125, 130, 267. See also Novelas on Spanish television Social class. See Class Social constructionism, 46-47 Social face. See Face Socialization of coolness in African Americans, 32 of gay males, 259-269 of gender identities, 84 self-managed socialization by gay teenagers, 264-266, 268-269 of women's language, 84 Sociolinguistics body in, 202-203 on codes witching, 389 data collection in, 163, 164-165 definition of, 29 on identity, 317, 318 sex as dichotomous variable in, 201

429 sex stereotyping in, 200-201 of Tunisia, 203-215 and "two cultures" model of language and gender, 201 Sounding, 32-33. See also Signifying Spanish, 66, 69, 273, 278, 282, 389, 392, 393, 396-397, 399, 402. See also Latinas/os Speech. See Conversation; Interaction; Language Stance, 48, 226, 273-274, 404n5 Standard English on the shopping channel, 355, 357 whites as speakers of, 29 and WOMAN AS DESSERT metaphor, 149 Status connection and, 227-234, 239n8 of Irish language, 372 of Tunisian women, 208, 211-215 See also Power Stereotypes of African Americans, 30, 317 gender stereotypes in stories by Latina/o primary-grade students, 67, 70-79 Goffman on, 224 stereotypic speech in folktales, 110 stereotypic speech in Lakhota language, 110114 of white middle-class women, 321-322 See also Ideologies Stigmatization, 263, 296-297, 305 Stories of agoraphobic woman, 85-97 of elderly Appalachian woman, 16-17, 245255 function of, 65 of self-transformation, 46-47 See also Coming out; Life stories and personal narratives; Stories by Latina/o primary-grade students; Storytelling; Women's narrative Stories by Latina/o primary-grade students border-crossing in, 79 characters in, 67-77 "free-writing" process for, 64, 66-67, 79 gender segregation in, 70, 79 goodness/badness in, 73-77, 78, 79 lack of closure in, 80-81n3 love and romance in, 71-74 poverty/wealth in, 77-78 power in, 77-78 processes and practices in writing, 66-67 social relations in, 69-80 superheroes in, 74-75 thematic tensions in, 70-71 Storytelling. See Narrative

430

Subjectivity "double subjectivity," 259 of life stories and personal narratives, 270n6 See also Identity; and headings beginning with Self Subversive versus misogynistic nature of drag queens, 315-317, 325-327 Swearing. See Profanity and vulgarity Syntax. See Grammar of Lakhota Talk. See Conversation; Interaction Teasing, 67, 261, 263 Teenage boys African American teenagers' use of signifying, 32-33, 35 coming-out experiences of gay teenagers, 17, 259-269 conversations of, in Australia, 141-142n2 resistance of working-class teens to regulatory forces of school, 79 suicide and, 269n3 voices of, in British teenage girls' conversations, 132 Teenage girls boys' voices in conversations of, 132 class differences in speech of, 124, 142n7 conversations of, in Australia, 141-142n2 discoursal range in conversations of, 125129, 132-141 friendship and femininity, 137-138 in gangs, 236 gender in conversations of British teens, 14, 123-41 instigating in African American speech, 3336, 37, 41 intertextuality in conversations of, 124, 129, 135-136 Latina teenagers' alignment and disalignment, 17, 273-291 linguistic features of conversations of, 133138 maternal voice in conversations of, 130-132 postfeminism and, 124-25 resistance of working-class teens to regulatory forces of school, 79 self-esteem of Latinas, 404n7 signifying in African Americans' speech, 33-36 suicide and, 404n7 swearing by, 132 turn-initial no among Latina adolescents, 273-291 voices in conversations of, 129-136

SUBJECT INDEX Teleshopping. See Home Shopping Network; Shopping channel Teletype communication device (TTY), 55, 56 Television. See Popular media Thai, 102 Themes of British teenage girls' conversations, 125141 of stories by Latina/o primary-grade students, 70-71 of women's narrative, 248, 252 Theory. See Feminist theory; Queer theory Therapy, 167 Throwing shade, 42n7 Topics. See Themes Transgender groups, 308n6, 313-314, 318, 321-322, 327nl, 344. See also Drag queens; Hijras Transgressivc identities, 9-10 Transsexuals, 313, 318, 321-322, 327nl, 344 Transvestites, 314, 327nl TTY, 55, 56 Tunica, 102 Tunisia diglossia in, 208-209 education in, 206, 214 gender and culture negotiation in, 15-16, 200-215 glottal stop in, 209-211 sociolinguistic situation in, 203-215 television in, 214-15 women's role and status in, 208, 211-215 Turn-initial no among Latina teenagers, 273291 Turn taking in British teenage girls' conversation, 134, 135, 136, 137 in conversation of Latina teenagers, 273-291 in girls' games, 397-399, 402 in hopscotch, 397-399 Vernacular of African Americans as male and poor, 2930 definition of, 30 reading dialect by African Americans, 36-37, 40 See also African American Vernacular English; Nonstandard language Voice boys' voices in British teenage girls' conversations, 132 in British teenage girls' conversations, 129136 of gay men, 268

SUBJECT INDEX maternal voice in British teenage girls' conversation, 130-132 multiple voices of women of color, 6 Vulgarity. Sec Profanity and vulgarity West Side Story, 261, 266 Whites. See British teenage girls; European Americans "Winter's King" (Le Guin), 336 WNAI. See Women's National Indian Association (WNAI) WOMAN AS DESSERT metaphor amelioration and, 150-152 evolution of, 146-149 lexical ambiguity and psycholinguistics, 156 lexical domain overlap and, 154 list of terms, 157-158 pejoration and, 150-152 phonetic considerations in, 154-156 phonosemantics and, 155-156 semantic shifts regarding, 150-152 and stereotype of promiscuity, 150-154 synchronic feature analysis of, 152-154 syntactic depersonalization and, 154 WOMAN AS OBJECT metaphors, 14, 146 Woman on the Edge of Time (Piercy), 334, 341343, 346n7 Women. See African Americans; European Americans; Feminism; Latinas/os; Native Americans; Women's narrative; and headings beginning with Gender, Woman, and Women Women of color feminism and, 4-7, 29 multiple identity positions of, 5, 6 See also African Americans; Latinas/os; Native Americans Women's language African American drag queens' use of, 321-327 in Appalachian woman's stories, 16-17, 245255 in Lakhota, 13-14, 101-117 models of women's narrative, 241-245, 254255, 255n3 prescriptive norm for, 317 sociali/ation of, 84 in workplace, 232-238 See also Women's narrative; and headings beginning with Gender

431

Women's movement. See Feminism Women's narrative behavioral features of, 250-251, 252-254 characteristics of, 243-244 formal and contextual features of, 248-250, 252-253 "kernel stories" and, 243, 255nl models of, 241-245, 254-255, 255n3 "personal-experience narratives" and identity, 242-243 self-presentation in, 252-253 themes of, 248, 252 Women's National Indian Association (WNAI), 184 Words of the Tyrtle, 166, 175-176, 177 Workplace discourse between men, 229-232, 377-383 between women, 232-233 rapport talk, 232-233 same-sex hierarchical office interactions, 16, 221-238 sex-class-linked framing and, 16, 222-227, 229-234, 237-238 status and connection in, 229-234, 239n8 World Wide Web. See Internet Writing. See Letters to boarding-school officials; Life stories and personal narratives; Literacy; Stories by Latina/o primary-grade students Yana, 102, 103, 105 Youth. See Children; Teenage boys; Teenage girls Zines compared with magazines, 169 compared with online journals, 176-177 discourse of, 168 fanzines, 169 form and function of, 165, 166, 167-170, 177-178, 178n4 as forum for metadiscourse, 172-174 history of, 169 marginalized populations and, 165 perzines, 170-172 revelation of the self in, 170-174, 177-178 violations of maxims and nonstandard forms in, 165, 168, 170-172